######general
2/10/2017 4:42 PM
sydneyneurotechx :
<!channel> Please help me in saying Happy Birthday to <@U073XJ5B6>. One of the cofounders and Executive Director of NeuroTechX.
2/10/2017 4:43 PM
alex :
🎂 🎂 🎂
2/10/2017 4:43 PM
sydneyneurotechx :
🎵
Happy Birthday to you,
Happy Birthday to you,
Happy Birthday dear Yannick,
Happpppppppppppppy Birthday tooooooo youuuuuuu
🎵
2/10/2017 4:43 PM
adam :
Happy Birthday Yannick! Thank you for everything you've done and continue to do for the organization. Hope you can take a day off and have some fun today!!
2/10/2017 4:44 PM
perrotta :
Happy Birthday <@U073XJ5B6>!
2/10/2017 4:44 PM
eferdinand :
Happy Birthday Yannick 🍰
2/10/2017 4:44 PM
nirmeshg :
Happy birthday yannick
2/10/2017 4:44 PM
davidevaleriani :
Happy birthday <@U073XJ5B6>, enjoy the spamming of the tags 😛
2/10/2017 4:46 PM
romy_lorenz :
besides neurotechX I am also passionanate about arts+neuroscience and work for the curatorial collective AXNS - we just launched a crowdfunding campaign to support our upcoming LSD brain wave signal and sound art hackathon. Check it out, spread the word and consider backing this exciting project: https://crowd.science/campaigns/lsd-brainwave-and-sound-art-hackathon/
2/10/2017 4:46 PM
endoxaneuro :
🎹 🎬🎤🎼🎼🎼🎼🎼🎻🎻🎻🎻🎷🎷🎷🎺🎺🎺🎸🎺🎸🎸 HAPPY BIRTHDAY YANNICK! You are lucky 🍀 to be born in February like me 😊 all my best wishes for an older greater year ahead of you 🎂🎂🎂🍰🍰🍰🍰
2/10/2017 5:42 PM
ch.yumin :
<@U073XJ5B6> Happy birthday!!!!
2/10/2017 5:44 PM
ray_cassani :
Hey <@U073XJ5B6>, happy birthday 🎂 !
2/10/2017 5:49 PM
tt :
Happy birthday <@U073XJ5B6>, thank you for your work and being awesome.
2/10/2017 5:58 PM
ryanlintott :
<@U073XJ5B6> Happy birthday!!
2/10/2017 6:00 PM
sidksv :
Happy birthday <@U073XJ5B6> !
2/10/2017 6:38 PM
stephen :
:ntx:
2/10/2017 6:38 PM
yannick :
<@U073Y8TBR> :ntx2: is nicer 🙂
2/10/2017 6:39 PM
yannick :
or :ntx3: 🙂
2/10/2017 6:39 PM
yannick :
:ntx3:
2/10/2017 6:43 PM
stephen :
:ntx2: :ntx3: :ntxt:
2/10/2017 6:47 PM
bryan_j :
Happy birthday Yannick! It has been a pleasure working with you over the last year and a half, and I look forward to many more opportunities to do so. Celebrate well this weekend!
2/10/2017 7:11 PM
sonia :
Happy birthday Yannick 🎈
2/10/2017 9:18 PM
lai_london :
Happy Birthday and many happy returns Yannick
2/10/2017 9:18 PM
lai_london :
(Your clearly loved!)
2/10/2017 9:19 PM
tudor :
Happy birthday Yannick!
2/10/2017 9:21 PM
dano :
Hey Happy Birthday, Yannick! Thanks for getting this crazy thing started
2/10/2017 9:29 PM
hish :
Happy Birthday Yannick.
👍
2/10/2017 10:59 PM
yixuan :
Happy birthday Yannick!! 🎂
2/11/2017 3:11 AM
yannick :
Thanks everyone for the wishes! I really appreciate it.
2/11/2017 2:49 PM
z3 :
Happy B-Day! :ntx3:🧠 🍻 x)
2/15/2017 9:02 PM
dano :
Does anyone have an Android tablet and a Muse and want to help the interactive EEG tutorial project? We need testers
2/15/2017 10:09 PM
adamm_neurable :
Anyone going to be at TechX in San Francisco this March? And/or in the Bay Area? I will be visiting that week 😁
2/15/2017 10:10 PM
yannick :
<@U0752QUB1> is going and he might know other people going.
2/15/2017 10:11 PM
jnaulty :
Ya! I'll be there
2/15/2017 10:11 PM
yannick :
<@U2XQDGG9J> are you guys having a booth or something?
2/15/2017 10:11 PM
adamm_neurable :
No, just browsing 😊
2/15/2017 10:11 PM
jnaulty :
Check out <#C0VTQK7CZ|_san-francisco> too
2/15/2017 10:11 PM
yannick :
You got your ticket already?
2/15/2017 10:12 PM
adamm_neurable :
Yes
2/15/2017 10:13 PM
yannick :
Ok! (because the code NeurotechX gets you a nice discount, was asking in case!)
2/20/2017 11:59 PM
bhargava2566302 :
Hey I'm attending the conference this year see you all guys
2/21/2017 2:43 AM
sabrina.k.tang :
Hey! Is there anyone else here from Halifax? Would love to get in touch.
2/21/2017 11:14 AM
firassafieddine :
hello everyone
I am currently trying to extract data from an epoc emotic 14 channel eeg headset , I am having some trouble being able to extract the data stream
Any suggestions?!
2/21/2017 11:50 AM
firassafieddine :
but yeah the raw data would do as well
2/21/2017 11:55 AM
nicolettebraingineers :
I�m not quite sure if the drivers from the SDK are available for Arduino
2/21/2017 12:01 PM
firassafieddine :
i would love to, i am not really sure my programming skills are enough, do you have any advice?
2/21/2017 12:06 PM
firassafieddine :
i use windows generally, but i have a linux machine too
2/21/2017 12:15 PM
nicolettebraingineers :
(If you want to program it yourself)
2/21/2017 12:20 PM
firassafieddine :
thanks for taking the time to help anyway 🙂
2/21/2017 12:24 PM
dlz :
I'm not sure how it is license-wise with the availability of the data, but we were able to relatively easily access the affective values from the suite (also, from the looks of your screenshot an older version...) with mindYourOSC's in processing, you should be able to just select these values in there (I used an emotiv with research license though, so we had access to all data on it, maybe that makes a difference)
2/21/2017 2:30 PM
jfrey :
Hi everyone, if you do research and are interested in social interactions and physiological sensors at large, you might have a look at a special session that will take place during ACII: "Brain and physiological signals for multi-user modeling". Here is the associated website: http://www.affective-sciences.org/en/bps-mum We would be glad to get your contributions, I am sure NeuroTechX folks around here do pretty amazing science 🙂
2/22/2017 6:09 PM
bryan_j :
Anyone in Toronto looking for a team to attend the 2017 DementiaHack? Message me ASAP! Thanks!
2/23/2017 4:22 PM
sydneyneurotechx :
http://spectrum.ieee.org/the-human-os/biomedical/imaging/ai-predicts-autism-from-infant-brain-scans
2/23/2017 11:01 PM
stephen :
Has anyone here used https://neurodata.io/?
2/23/2017 11:52 PM
bciguy :
<@U073Y8TBR> great find - never heard of this project.
2/28/2017 1:09 PM
bhargava2566302 :
Thats cool
2/28/2017 1:18 PM
yannick :
Hey, I'm currently working on a presentation about the latest news & breakthroughs in NeuroTech, focusing on 2016 (kind of a review). We'd like for the different NTX chapters to have access to it and do an event to showcase what's new in the field. From "wow" kinda news to scientific breakthroughs, to new kickstarters, to new laws, to major investments, etc.
The goal of the presentation (and the event) is that when you leave, you feel like you have the big picture of what happened lately - without digging too much into the science of it, but staying at a higher level.
And obviously you'd leave inspired, because the field is awesome 😛
- Do you know any review already online that could be useful?
- What would you put in that review/presentation? (what for you, is worth mentioning that happened in 2016)
- If you'd like to help with the content and/or the presentation, please PM me 🙂
Feel free to throw anything that comes to mind!
(obviously the presentation will be accessible/shared)
3/1/2017 2:38 AM
jacobflood :
Hey everyone, big news: Mindset�s Kickstarter goes live tomorrow morning at 8am EST!
For those of you who aren�t familiar, Mindset is a pair of overear headphones that integrate EEG sensors into the band to measure and improve concentration. They�re bluetooth, noise cancelling, and give you full access to the data. You can get more info at http://thinkmindset.com|thinkmindset.com, which will also redirect to the Kickstarter once it�s up.
The campaign goes live tomorrow at 8am EST. Be the first to grab one of the early birds, and get 46% off! 😄
3/1/2017 12:44 PM
lai_london :
Mindset website looks remarkably impressive
3/1/2017 1:01 PM
jacobflood :
MINDSET�S KICKSTARTER IS LIVE!!! 😄
Here�s the link you can use to back the project: https://www.kickstarter.com/projects/mindset/headphones
Feel free to give me a shout at 438 838 1477 if you need help, or check out this video explaining how to back kickstarter projects http://bit.ly/2msDIdZ.
Thank you so much for your help!
3/1/2017 5:31 PM
eferdinand :
We're proud to partner w/ #OReillyAI in New York, NY June 26-29. Save 20% w/code UGNTMTL
3/1/2017 6:05 PM
yannick :
Hahaha! I still prefer :ntx2: :ntx3: :ntxt:
3/1/2017 7:40 PM
graeme :
http://journal.frontiersin.org/article/10.3389/fnins.2017.00109/abstract
3/1/2017 8:03 PM
yannick :
:Muse:
3/1/2017 9:36 PM
sc :
jacobflood: Congrats on the Kickstarter launch! One question though, did you research the trademark on the name "MindSet"? Neurosky's first consumer EEG (in 2009) was also called the MindSet, and was based around an audio headphone design. A quick Google turns up some articles and even the manual which feature a (TM) by the name.
3/1/2017 9:46 PM
jacobflood :
Thanks for the support!
We�re aware, and as we understand the product is discontinued, and not currently under trademark. Could you send us the trademark manual you found?
3/1/2017 10:03 PM
sc :
Sure, and good luck with it all, just something which stood out as soon as spotted. In any case this manual has the (TM) symbol at least once http://developer.neurosky.com/docs/doku.php?id=mindset_instruction_manual and this Reseller Agreementon the Necomimi site lists MindSet as a NeuroSky trademark http://store.necomimi.com/pages/reseller-agreement
3/1/2017 10:20 PM
jacobflood :
Good to know! I�d seen that as well, but when we did a formal trademark search we didn�t find anything filed. We�re concluding the filing in the next few days
3/2/2017 11:54 PM
dano :
A babe is born: the Neurodoro app, coming out of Toronto!!
https://github.com/jdpigeon/neurodoro
3/3/2017 7:54 PM
bhargava2566302 :
Can you explain it purpose
3/3/2017 11:28 PM
martinperez :
hello , I'm interested in contacting someone from Peru
3/3/2017 11:43 PM
yannick :
Hey martinperez you should join <#C0VU2AP1Q|_lima> and talk with the folks there 🙂
3/4/2017 12:16 AM
alexandre.barachant :
yannick do you have insider information ? i will be in NYC the 28. i'm intested to know more
3/4/2017 12:17 AM
yannick :
Of course I do 😉
3/4/2017 12:17 AM
yrenard :
yannick has all information and more
3/4/2017 12:17 AM
yrenard :
we should make a bot of you yannick
3/6/2017 1:27 AM
dan :
Startup Weekend: Maker Edition is this weekend coming up. Would anyone consider making the trip and starting an impactful organization?
3/6/2017 3:14 AM
c00p3r :
question will anyone be at SXSW this year in austin texas?
3/6/2017 3:27 AM
graeme :
yannick what are those headsets in the mindgamers promo thing?
3/6/2017 8:01 PM
pharo :
For those that are in San Francisco I will be visiting from March 13-17th. It would be great to connect with other Neuro Tech X people.
3/6/2017 8:05 PM
sydneyneurotechx :
pharo . I'd suggest posting in the <#C0VTQK7CZ|_san-francisco> channel that you will be around
3/6/2017 9:38 PM
pharo :
Sounds great thank you for directing me to the right place
3/7/2017 2:03 AM
yannick :
graeme it's a new / custom made Cognionics device that will be used for the EEG recording.
3/7/2017 1:34 PM
alexandre.barachant :
yannick do they plan to add it to their catalog ?
3/7/2017 1:38 PM
yannick :
Not sure of the state of the "product", if they'll just use it themselves for that kind of projects/collaboration or make it available for purchase.
3/7/2017 7:02 PM
alainafinley :
Hi there! My name is Alaina Finley and I am currently an undergrad student pursuing a Computer Science degree at the University of Utah. This past summer I worked as an intern for Push the World in NYC ( aj ) I definitively want to go into the NeuroTech industry as neuroscience is my passion, however with the courses I�m taking in CS, I�m only exposed to neuroscience via independently reading research papers. (I�m going into CS so I can eventually go into neurotech..and I�m really feeling disconnected to the field) My school offers a create your own degree program that I know many kids have gone through and have had success, and I was thinking of switching and doing an Applied Neuroscience degree with an emphasis in computing. I was wondering if anyone has any input in making this move? And any advice (especially to courses you�d recommend I take)?
3/7/2017 7:51 PM
w :
Does anyone here use Research Gate?
3/7/2017 7:53 PM
yrenard :
Does having an account and asking to reset forgotten password from time to time count as using Research Gate ?
3/7/2017 7:54 PM
yrenard :
Haven't found much 'usefullness' about it beyond being notified on peers work - I think scholar would even make a better job at it - What are you looking for ?
3/7/2017 8:00 PM
w :
I can't get an account unless I'm affiliated with an institution or company that does research. I'm not. But after the whole process of applying as an individual, they put you on a wait list and say you can get off of it by giving them the account emails of people you know that use it. Ugh.
3/7/2017 8:02 PM
w :
They apparently don't even notify the people you list. So why do they need to know and how do they use that information?
3/7/2017 8:02 PM
w :
¯_(Ä)_/¯
3/7/2017 8:03 PM
w :
Maybe they're graphing the research landscape - who's tied to who and who's researching what - and finding the edges of that
3/7/2017 8:03 PM
yannick :
You can use me to get access.
3/7/2017 8:06 PM
yrenard :
you can use me as well
3/7/2017 8:07 PM
yrenard :
I was looking if I could invite you but looks like this is not a feature
3/7/2017 8:07 PM
yrenard :
oh yes, I found it !
3/7/2017 8:08 PM
yannick :
Refer a friend and get a +1 publication 🙂
3/7/2017 8:08 PM
yannick :
(a "private" publication :p)
3/7/2017 9:33 PM
sydneyneurotechx :
alainafinley: Hey alainafinley The answer to your question will depend on what your interests are and what you want to specialize in. Neurotech is a multi disciplinary field and you can contribute to it in a variety of different ways. If you want to do research, ask yourself what element of Neuroscience you would like to get involved in. In the meantime, I'd suggest looking at some Neuroscience Moocs to compliment your CS studies and take a look at these resources: https://github.com/NeuroTechX/awesome-bci The one which may interest you the most is the analyzing neural time series book. I'd also suggest talking to dano and see about contributing to the <#C0K883P71|interactive-tutorial> project https://mitpress.mit.edu/books/analyzing-neural-time-series-data
3/8/2017 1:29 AM
z3 :
https://qz.com/866352/scientists-say-your-mind-isnt-confined-to-your-brain-or-even-your-body/
3/8/2017 1:50 AM
z3 :
sry if this fall more in <#C073WQK34|random>
3/8/2017 6:17 AM
gabe.ibagon :
hey alainafinley! in terms of putting together courses for a degree, you should check out the degree requirements for a neuroscience major at some undergraduate colleges, and see what classes they generally suggest (for example, i think Brown has a good program: https://www.brown.edu/academics/neuroscience/undergraduate/neuroscience-concentration-requirements). Usually starts off with some introductory bio, chem, physics, psych, and then assorted advanced neuroscience courses, but it's dependent on the style of the degree and what you want to focus on.
3/8/2017 7:27 AM
physionikhil :
It great
3/8/2017 4:17 PM
jfrey :
Hey! If you attend to the CHI conference that will take place in Denver in May ( https://chi2017.acm.org/ ), we will give courses about... BCIs. It's a 3-parts class, with 1. Introduction to the domain (targeted at HCI researchers); 2. Hands on session with OpenViBE; 3. Pitfalls & ethics. See https://team.inria.fr/potioc/bci-courses/ . Even if you don't participate to these (awesome) courses, I hope I will see some of you there; last year I was pleasantly surprise to meet NeuroTechX pairs 🙂
3/8/2017 4:35 PM
dojeda :
To whom it may concern : a hack to put latex on Slack: https://github.com/fsavje/math-with-slack...
it's a client side though... so if your slacker mathematician best friend does not have it, he will not see it
3/8/2017 4:36 PM
dojeda :
You see
3/8/2017 4:39 PM
alexandre.barachant :
awesome, i can type math now !
3/8/2017 4:45 PM
dojeda :
I am not sure why it does not work out-of-the-box... I tried yesterday and did not work, now I tried again and it works
3/8/2017 4:45 PM
dojeda :
¯_(Ä)_/¯
3/8/2017 4:52 PM
alexandre.barachant :
you have to restart completely slack
3/8/2017 5:12 PM
yrenard :
does it survive to auto-updates ?
3/8/2017 5:15 PM
dojeda :
most certainly not
3/8/2017 5:19 PM
sydneyneurotechx :
dojeda Nice work! Would be nice to have an app which we could integrate into Slack
3/8/2017 5:20 PM
qbarthelemy :
$$ N^{EURO} T_{ECH} X $$
3/8/2017 5:21 PM
sydneyneurotechx :
I know there are research labs out there that use slack as their communication method, so I can definitely see a market for it
3/8/2017 5:25 PM
dojeda :
Well, if you want a bot that does that, it would have to watch for some expressions, and then inject a JS script that does the magic... I don't know if it's possible for the clients to receive anything other than text
3/8/2017 5:26 PM
dojeda :
BTW this is not a hack of my own, I just found it yesterday on github...
3/8/2017 5:27 PM
sydneyneurotechx :
dojeda: Ah ok. Good find nonetheless ^_^
3/8/2017 5:30 PM
amine.abdessemed :
The client can receive images like the ones you uploaded above.
3/8/2017 5:33 PM
dojeda :
But the mathjax hack here is that mathjax replaces "dollar latex dollar" with a <span> with the rendered output
3/8/2017 5:36 PM
amine.abdessemed :
You're right, if we add the feature to the bot, he will generate a second rendered message, spamy but the advantage is, it would be visible to everyone and is probably update safe.
3/8/2017 5:36 PM
qbarthelemy :
test : $ N^{\kern-0.25emEURO}\kern-0.25emT_{\kern-0.15emECH}\kern-0.25emX $
3/8/2017 5:37 PM
qbarthelemy :
works for me!
3/8/2017 5:48 PM
sydneyneurotechx :
amine.abdessemed is it possible for the second rendered message to be added to thread of the original message instead of being a standalone one?
3/8/2017 5:48 PM
amine.abdessemed :
yeah
3/8/2017 6:18 PM
dojeda :
but that would create a thread (or a new message)... wouldn't it be better to replace the message? I think it's possible but I'm not very familiar with the bot API
3/8/2017 6:19 PM
dano :
Hey alainafinley, I might also suggest looking in to the cognitive science minor at your school. The coursework will overlap with your program in CS and, in my experience, professors in cognitive science are usually the most likely to expose students to bleeding-edge neurotech/AI stuff
http://philosophy.utah.edu/undergraduate/cognitive-science.php
3/8/2017 6:27 PM
dano :
I'm on there -> Dano Morrison
3/8/2017 6:35 PM
sydneyneurotechx :
Apparently it's not possible via the API according to Amine.
3/11/2017 4:36 AM
micslab :
Hey folks has anyone done any product viability tests for a wearable that helps rock climbers stay in a state of flow/ mindfulness with slight beta expression?
3/13/2017 11:13 PM
sydneyneurotechx :
To view archived text from Slack please visit:https://github.com/NeuroTechX/ntx_slack_archive/blob/master/general.md
3/14/2017 10:22 AM
fsuarezj :
Hi everybody,
I'm a computer engineer interested in neurotechnology, so I'm learning about it and planning to buy a device within the next months. But before choosing what device to buy I would like to learn about what I want to do, to make the best choice. I would like to research about pain detection and/or mood state detection. Do you know any resources or other papers about it?
Thanks a lot!!
3/14/2017 2:55 PM
friedstoeren :
Hi there. I have question: same days ago I saw a documentation about chips that get implemented to the arm and then you can robotarm and so on. It there such a technic and how is the name of it?
I'm really interested to get one of then.
3/14/2017 3:03 PM
yannick :
"chips that get implemented to the arm and then you can robotarm and so on"? Not sure I understand what you mean. Are you talking about an invasive technology (implanted / under the skin)? Or non-invasive technology like EMG reading muscle activity allowing you to send commands to control 'something'.
3/14/2017 3:14 PM
friedstoeren :
yannick: in this documentation it was invasive
3/14/2017 3:15 PM
friedstoeren :
Sorry for my bad English... I was not writing and talking English for years.
3/14/2017 4:27 PM
graeme :
The full Muse ERP & P300 paper has been published.
3/14/2017 4:27 PM
graeme :
http://journal.frontiersin.org/article/10.3389/fnins.2017.00109/full
3/14/2017 4:28 PM
graeme :
If you're looking for the matlab code to replicate that experiment, you can find it at http://www.neuroeconlab.com/muse.html
3/14/2017 4:29 PM
graeme :
We'll be demonstrating this and NeuroTechTO's EEG 101 at the Rotman Research Conference next week, so if you're in attendance, please stop by and check it out
3/14/2017 4:29 PM
graeme :
3/15/2017 10:45 AM
omer :
not sure how you can deal with the excessive sweat
3/16/2017 2:52 AM
aj :
Awesome read! Good work!
3/16/2017 4:28 PM
yannick :
Frontiers in Neuroscience now follows NeuroTechX on Twitter.
(Achievement Unlocked I guess :p)
3/16/2017 9:56 PM
alainafinley :
This is awesome! Thank you all so much for the insight.
3/17/2017 12:28 AM
dano :
wuddup leonfrench
3/17/2017 12:29 AM
dano :
First Toronto hacknight at VarageSale nearly confirmed...
3/17/2017 3:53 PM
leonfrench :
Hello and thanks for the invite
3/18/2017 11:08 PM
mkos :
hi - I have a question about Muse 2016 - is this the right place to ask?
3/19/2017 2:16 AM
graeme :
yannick: you reminded me that I think I still have the password for that Twitter account. Who else should it follow? 😁
3/19/2017 2:17 AM
graeme :
mkos send a direct message to me or hubertjb
3/21/2017 11:40 AM
yannick :
Hey teon & alexandre.barachant - if people are interested in the "Supervised Neural Time Series 2017 sprint" (March 27th - 31st, in NYC) - https://kingjr.github.io/supervised_time_series/ - what's the deadline to register?
I wish I could be there, the event sounds nice!
3/21/2017 12:03 PM
igweckay :
Legit! Was there a register link on the site? I didn't find one
3/21/2017 12:09 PM
alexandre.barachant :
Jean-remi king is the organizer of the event, you might want to contact him : https://sites.google.com/site/jeanremiking/
3/21/2017 5:01 PM
aj :
I'm gonna go!
3/21/2017 6:53 PM
mkos :
looks like http://neurotechx.com|neurotechx.com is down 😕
3/21/2017 7:36 PM
yannick :
Is it? We changed the hosting & DNS, so it might be a "cache" issue. Do you mind clearing your cache and/or trying another browser (if you have).
3/21/2017 7:57 PM
benjamindeleener :
Works for me!
3/21/2017 8:43 PM
dano :
it's up!
3/22/2017 5:50 AM
w :
Yves Meyer, Wavelet Expert, Wins Abel Prize
http://hn.premii.com/#/article/13921208
3/22/2017 4:59 PM
mkos :
Interesting, I keep getting "ERR_CONNECTION_REFUSED"
3/22/2017 5:00 PM
mkos :
I tried Chrome and Firefox
3/22/2017 5:02 PM
mkos :
Also, a few days ago it said that the website was hacked...
3/22/2017 5:15 PM
mertnesvat :
Hello everyone I�m new but eager to the topic and I have general question.
I�m human computer interaction student at tallinn. I�ve recently discovered this workshop which will be take in place at Cambridge, UK since I�m not living UK I calculated all the cost transportation+accommodation etc and it�s around 850 ¬ and I�m confused does it worth it or not.. I can afford it but as all students my budget is limited and want to spend it carefully 🙂
3/22/2017 5:15 PM
mertnesvat :
3/22/2017 5:33 PM
davidevaleriani :
I think it covers too many topics for four days, so you will probably end up knowing a bit of everything. For getting that kind of knowledge, you can check online resources, such as our http://www.neurotechedu.com|www.neurotechedu.com, or read some good books like these https://github.com/NeuroTechX/awesome-bci#technical-books or the Introduction to Event-Related Potentials by Luck.
3/22/2017 5:50 PM
sydneyneurotechx :
mertnesvat . Going to an event like that will be great to get connected with people, however if you are looking for knowledge, there are some good online resources available as davidevaleriani mentioned
3/22/2017 7:06 PM
mertnesvat :
Thanks davidevaleriani yes it seems good start point 👍
3/22/2017 7:25 PM
benjamindeleener :
I think it is a problem with the http/https security
3/22/2017 7:25 PM
benjamindeleener :
you need to allow the website to be opened, in the same page it was refused
3/24/2017 3:03 AM
sydneyneurotechx :
congrats on the mit tech review article pharo https://www.technologyreview.com/s/603896/controlling-vr-with-your-mind/
3/24/2017 11:52 PM
eferdinand :
Call for speakers -- O'Reilly Artificial Intelligence Conference in San Francisco
The O�Reilly AI Conference is coming to San Francisco September 17-20, 2017 to explore the most essential and intriguing topics in intelligence engineering and applied AI. We�re looking for compelling case studies, technical sessions, tear-downs of both successful and failed AI projects, technical and organizational best practices, and more. Apply to speak by March 28.
http://www.oreilly.com/pub/cpc/68827
3/27/2017 10:25 PM
sydneyneurotechx :
http://www.theverge.com/2017/3/27/15077864/elon-musk-neuralink-brain-computer-interface-ai-cyborgs
3/28/2017 1:01 AM
sc :
ha you did beat me to posting that. I can never tell if <#C073WP1T9|general> is the right place or <#C073WQK34|random> for articles. And if anyone decides to make a #news channel, then I'll never know which of the 3 to use (c:
3/28/2017 1:02 AM
dano :
Definitely general worthy. Let's wait and see if they put out a call for hires
3/28/2017 1:03 AM
dano :
Also, some good BCI discussion in the Hacker News comments + a Muse mention
https://news.ycombinator.com/item?id=13971728#13972913
3/28/2017 3:52 AM
sydneyneurotechx :
sc haha yea, I figured it was going to be a race between me, yannick, you and stephen in posting it here. There is a <#C0HBD8S86|neurotechnews> so maybe we can post in there.
3/28/2017 5:10 AM
physionikhil :
Neuralink doesn't have a website as yet
3/28/2017 3:08 PM
sydneyneurotechx :
3/28/2017 3:16 PM
sydneyneurotechx :
<!channel> ^
3/28/2017 4:42 PM
physionikhil :
Thnx , they r only hiring for the time being
3/28/2017 6:43 PM
adamm_neurable :
It's already begun...http://bostinno.streetwise.co/2017/03/28/elon-musks-neuralink-hires-boston-university-researcher/?mc_cid=4e967feddd&mc_eid=172080915b
3/28/2017 7:14 PM
sydneyneurotechx :
sc I think you can take credit for the Neural Lace idea 😛
3/28/2017 9:59 PM
stephen :
It�s really exciting that Elon is joining the neurotechnology industry
3/28/2017 10:00 PM
stephen :
It�s really exciting that Elon is joining the neurotechnology industry
3/29/2017 1:28 AM
a.tech :
Yeaaa! I bet you that because of this news fund raising will be a little easier for neurotech startups looking forward
3/29/2017 7:20 AM
sydneyneurotechx :
Fellow Neurotecher melanie recently had her article released for IEEE on DBS in the military. Part of the article provides a good perspective as to whether cognitive implants are in fact worth it. http://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7879372
3/29/2017 7:21 AM
sydneyneurotechx :
"These advances in medical technology
are inextricably linked to discussions
of human augmentation,
especially in the context of military
research. However, the existence
of �super-soldiers� enhanced by
brain implants are far from a reality.
Although a great deal of progress
has been made in animal models
towards aims such as accelerated
learning and increased memory
capacity using implantable brain
technology, research in humans
is currently limited to therapeutic
applications. Even if the therapeutic
applications were to augment function
beyond current human capabilities,
it is still an open question
whether or not the risks would be
deemed acceptable for non-clinical
populations. The only precedent
for cognitive enhancement comes
from pharmaceuticals, which are
often used to keep soldiers alert.
Similarly, non-invasive brain stimulation
(transcranial direct current
stimulation (tDCS) and transcranial
magnetic stimulation (TMS)) techniques
are being investigated for
accelerated learning, improved decision
making, and alertness. However,
unlike an implanted device,
these substances and techniques
can be used on a per-need basis.
While they do carry risks, both long
and short term, their use can be discontinued
once they are no longer
needed. A brain implant, although
considered reversible, carries the
risks of surgery at both implantation
and at removal, and cannot
be replaced or repaired outside of
a hospital"
3/29/2017 8:29 AM
tenoke :
Literally every article on the subject has a quote like that. Yes, we dont have much outside of theurapetic applications today, which is why we want to develop that for tomorrow. What's your point?
3/29/2017 2:45 PM
karchie :
there's risk in unsupported hype. around neurotech (interpreting the term very broadly), there are relatively easy problems (stimulating/reading anything peripheral) and moderately easy problems (deep brain stimulation), and then there's anything involving neocortex. with pyramidal cells, there's almost certainly semantic content not just to the coarse location and intensity of stimulation, but to the fine location (certainly apical vs. basal dendrites, maybe even at the level of individual branches) and the fine temporal structure; and these details will vary from one cell to the next. I've been hearing about cortical prosthetic modules (for a particularly egregious example) for 20 years; but if that's going to work at all, I think at best there's a period of months while the nearby neurons learn how to interface with this chunk of new stuff. That's without even getting into the biochemical problems like controlling inflammation and infection and creating a local chemical environment to promote plasticity. I mean, yes, march on with the rats, but I'll bet a nickel against any sort of cognitive upgrades in my lifetime. (sensory upgrades, sure, but noninvasive approaches work pretty well there.)
3/29/2017 2:47 PM
karchie :
(sorry. professional storm cloud is right there on my resume)
3/29/2017 3:14 PM
jmhorschig :
There was a report of Nuffield Council on Bioethics on this matter in 2013. Their report is extremely elaborate on that issue: https://nuffieldbioethics.org/wp-content/uploads/2013/06/Novel_neurotechnologies_report_PDF_web_0.pdf (and while on it, also check out https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4072086/)
3/29/2017 3:23 PM
karchie :
jmhorschig wow, that's great, thanks. haven't read in detail yet but at a glance it looks like a really nice summary worth a careful read.
3/29/2017 3:32 PM
yannick :
Thanks for sharing jmhorschig indeed it looks very interesting and worth a careful read!
3/29/2017 5:22 PM
yannick :
Another interesting piece, posted following Elon Musk news
"So it�s important to keep all of these considerations in mind when discussing Neuralink, Kernel, and the scores of brain-computer interface companies that will undoubtedly sprout up in their wake. It�s fun to talk about fighting Skynet or preventing The Matrix, but Musk and other entrepreneurs in Silicon Valley do have more realistic concerns and are, at least right now, more open to admitting that the more near-term goal is combating diseases and funding neuroscience research. It doesn�t grab headlines quite like combatting the extinction of the human race, but we should all rest easy knowing that we�re not quite there yet."
Elon Musk�s Neuralink is not about preventing an AI apocalypse
http://www.theverge.com/2017/3/28/15094040/elon-musk-neuralink-kernel-brain-chips-ai-apocalypse
3/29/2017 6:05 PM
amilenkovic :
I appreciate that this is getting coverage in general, and the links are great
3/29/2017 6:05 PM
amilenkovic :
the tone of the authors writing these articles really boggles me sometimes. It's so glib, talking about robot uprisings like it's a real thing around the corner
3/29/2017 6:05 PM
amilenkovic :
then the next article mocks the same ideas "ha ha no the robots aren't coming"
3/29/2017 6:05 PM
amilenkovic :
And lots and lots of FUD and how nothing will ever come of it or be viable
3/29/2017 6:06 PM
amilenkovic :
I just find it odd that it's all in there in a sort of smorgasbord of doubt and maybes and jokes, when the stuff they cover is very real and on the horizon
3/31/2017 3:42 PM
rohit :
http://news.mit.edu/2017/multifunctional-tiny-fibers-brain-0221
3/31/2017 4:12 PM
davidevaleriani :
Another piece related to Elon Musk with a NTX citation ;) https://theconversation.com/elon-musk-wants-to-merge-man-and-machine-heres-what-hell-need-to-work-out-75321
3/31/2017 4:13 PM
marvinandujar :
https://www.linkedin.com/pulse/thoughts-from-bci-researcher-elon-musks-new-company-marvin-andujar
4/2/2017 4:42 PM
d.adamos :
4/2/2017 6:57 PM
ray_cassani :
Hey everyone, good news, the BCI workshop repository has been updated. Before the workshop only ran Python2, now the scripts are Python3-compatible, and the same exercises are provided as MATLAB/Octave compatible scripts.
https://github.com/NeuroTechX/bci-workshop
4/2/2017 7:27 PM
dano :
d.adamos: would be interested in working on an open source implementation of their algorithm
4/2/2017 11:39 PM
sydneyneurotechx :
ray any plans to make Mules compatible with Mac OS?
4/3/2017 12:54 AM
benjamindeleener :
That would be nice
4/3/2017 1:03 AM
ray_cassani :
sydneyneurotechx, last year I made some tests, there isn't out-of-the-box support for Bluetooth in LabVIEW (2013, the license I have access) on OS X 10; the Muse support seems to have no-problem. I need to fix some small details to make the entire project fully Windows/OSX compatible rather than two different versions.
4/3/2017 2:15 AM
stephen :
We should make a blog post for bright-eyed, bushy-tailed engineers who discover EEG or tDCS, order a kit, and start making random things while caught up in the romance of �neurohacking.�
(Eventually these people tend to keep iterating on the same BCI paradigms, or leave neurotech entirely because those paradigms aren�t powerful and/or accessible enough to do actually new things.)
The post would describe the standard types of projects that have been done so far, and then list NEW types of projects that merit more attention from talented engineers.
yannick melanie brian.westerman andreacoravos a.tech
4/3/2017 3:05 AM
stephen :
Here�s a rough draft:
SO WHAT NOW, PARTY POOPER
Most of the interesting problems haven�t been solved yet.
What are problems that noninvasive neurotechnology developers face? How could we &?
- record higher-quality data
- record more types of data
- with sensors that are less invasive, smaller, and cheaper
Here are some specific projects that could be interesting, or problems to solve.
- make EEG, EMG, HRV, GSR, and IR sensors that are less invasive, smaller cheaper. (the ideal externally-worn ephys device would be invisible, weigh <1g, and cost <$1. how close can you get to that?)
- could you make temporary �sticker tattoos� that stick in the skin, transmit to an external device, and could be washed off with special solution? What if you just used normal tattoo-injection material in the scalp?
- connect all those tiny ephys things in a mesh network that also knows about your environment.
- data integration. h/t to Cloudbrain for starting on this.
- super-easy API to connect a game engine (Unity; Unreal&) with any ephys sensor.
- fNIRS. Hats off to <#C0TK3ULLS|openfnir> for a noble attempt. There are probably ways to do this even better and cheaper.
- create a way to detect �mental energy levels� throughout the day, with a device anyone could wear to any workplace. not HRV, not number of steps walked. Actual mental energy. Probably the most important physiologically-measurable variable for the lifestyle and careers that most people reading this post want. You could correlate this with sleep, diet, and anything else. I�ll start by offering a $1000 bounty for a good solution to this problem.
- find a way to help people to reliably have restorative sleep. This researcher found a way to turn REM sleep on and off with optogenetics (http://news.berkeley.edu/2015/10/15/researchers-find-neural-switch-that-turns-dreams-on-and-off/). Could you do that non-invasively?
4/3/2017 3:05 AM
stephen :
Here�s a rough draft:
4/3/2017 3:05 AM
stephen :
Here�s a rough draft:
PONG
There�s a pattern that�s common in neurotech:
Some enthusiastic engineers will discover EEG or tDCS, order a kit, and start making random things while caught up in the romance of �neurohacking.�
There are a couple of common forms this initial exploration takes:
- �A meditation app." "We�ll do neurofeedback to show people when they�re in a meditative state, and use that help them meditate more! Then more people will start meditating, and the world will become more conscious."
This is the �hello world� of neurohacking. A good idea for your own education, but Interaxon�s Muse is already a highly optimized version of this product. http://www.choosemuse.com/people-behind-muse/
1b) A variant: �Attention-tracking app.� People will wear an EEG at their computers and it will tell them when their focused!
To test whether people will actually use this, try offering a friend $20 to wear an EEG headset for six hours per day for one week.
- �Control X with my brain.� A mood-tracking LED shirt, a robot, your Spotify music playlist, a flamethrower, a drone. With motor imagery, you can use the brain as a up-down or left-right switch for any device. I wouldn�t say �don�t do this�. I would just say �be aware you�re doing this for your own education, in the way that freshman CS majors are often given homework to write their own version of Pong.� Because of the benefit-intrusiveness tradeoff of consumer EEG headsets, basically no one will walk around with an EEG on their head in order to express their emotions through your cool LED shirt. Heck, most people won�t consistently wear a fitness tracking bracelet (from which you can infer mood). (http://insights.dice.com/2017/02/07/lessons-jawbone-failures/).
- �tDCS�. "We heard on that one NPR episode that the army uses these to speed up learning. What if we could enhance learning?" That�s what Thync said in their original sales pitch.
Consider the first generation (2005-2017+) of consumer EEG and tDCS devices. Much of their sales momentum came from the novelty of a �brain-reading� or a �brain-zapping� device. After consumers order them, most are not consistently used because they were a wiz-bang gadget to show at a party, or an exciting thing to hack on, rather than consistently solving a problem that people have.
http://Foc.us|Foc.us recently introduced a product line targeted at DIY hackers: https://www.foc.us/eeg. Out of all the potential markets they could target, it�s alarming that brain hacking enthusiast are the largest.
If consumer neurotech wants to make really useful and commercially successful devices, we need to expand our technological toolkit.
4/3/2017 3:05 AM
stephen :
SO WHAT NOW, PARTY POOPER
Most of the interesting problems haven�t been solved yet.
What are problems that noninvasive neurotechnology developers face? How could we &?
- record higher-quality data
- record more types of data
- with sensors that are less invasive, smaller, and cheaper
Here are some specific projects that could be interesting, or problems to solve.
- make EEG, EMG, HRV, GSR, and IR sensors that are less invasive, smaller cheaper. (the ideal externally-worn ephys device would be invisible, weigh <1g, and cost <$1. how close can you get to that?)
- could you make temporary �sticker tattoos� that stick in the skin, transmit to an external device, and could be washed off with special solution? What if you just used normal tattoo-injection material in the scalp?
- connect all those tiny ephys things in a mesh network that also knows about your environment.
- data integration. h/t to Cloudbrain for starting on this.
- super-easy API to connect a game engine (Unity; Unreal&) with any ephys sensor.
- fNIRS. Hats off to <#C0TK3ULLS|openfnir> for a noble attempt. There are probably ways to do this even better and cheaper.
- create a way to detect �mental energy levels� throughout the day, with a device anyone could wear to any workplace. not HRV, not number of steps walked. Actual mental energy. Probably the most important physiologically-measurable variable for the lifestyle and careers that most people reading this post want. You could correlate this with sleep, diet, and anything else. I�ll start by offering a $1000 bounty for a good solution to this problem.
- find a way to help people to reliably have restorative sleep. This researcher found a way to turn REM sleep on and off with optogenetics (http://news.berkeley.edu/2015/10/15/researchers-find-neural-switch-that-turns-dreams-on-and-off/). Could you do that non-invasively?
4/3/2017 12:56 PM
karchie :
I really like this. It captures a lot of the skepticism I have about the field but also gives some concrete steps forward. I'm hoping to steal some time for a detailed review later this week but too many balls in the air so don't hold your breath.
4/3/2017 4:26 PM
stephen :
Thanks karchie would love your feedback
4/3/2017 5:28 PM
pierre :
stephen I agree with your post. One thing is that a lot of the solutions that you list would involve developing new technology, which is a solid step forward, but a hard problem for hobbyists and student clubs. Developing hardware takes decent funding and continued interest and investment from team members.
Beyond developing new technology, the other side is taking more advantage of the low-hanging fruit present with the current technology. Here are some recommendations for doing that:
- Read the literature! Scientists have been correlating various biosignals for decades and wrote about many things beyond motor imagery and meditation. A few examples are memory encoding, visual/auditory/sensory perception, emotion valence, various specific cognitive processing, etc. It will also give you a sense of what the current limits of the field are. If scientists can't get motor imagery to work reliably (>95% accuracy, which is roughly what would be needed for using an input device with low frustration) with 32 electrodes, then what's the chance of getting it to work with 4-8 electrodes?
- explore other avenues besides quantified self: things like diagnosis, drug detection, lie detection, "real" attention tracking for optimizing content, which involve using a device for a relatively short period of time, have been relatively underexplored in Neurotech, but are some of the areas where it has the most potential with the current tools
4/3/2017 5:36 PM
pierre :
As an addendum to "read the literature", you can get a lot of mileage when getting started from replicating scientific papers using biosignals to measure the effect you're curious about.
4/3/2017 5:38 PM
pierre :
Also, I think the current barrier to entry to all of this is that people don't know how to start recording & processing biosignals, and don't have a good sense of the whole field. <#C3525BTBN|neurotechedu> should help out with both of these problems
4/3/2017 5:40 PM
maxim :
Piling up on this discussion - what would you say are the limitations of what you can do with Muse?
4/3/2017 5:40 PM
maxim :
Is Calm app the maximum of what can be achieved with the little signal the device produces?
4/3/2017 5:57 PM
maxim :
Also, what are you talking about when saying "remix normal data science tools to create better workflows for dealing with neural data"? Any specific use cases in mind?
4/3/2017 6:05 PM
sydneyneurotechx :
In terms of hardware specs, it's comparable to most eeg systems. The main limitation is you are stuck to those electrode locations. It is a pretty well designed eeg. ERP and Oscillations are both possible with the Muse headband. Someone was also able to do Frontal Assymetry as well with the headband (https://imotions.com/blog/frontal-asymmetry-101-get-insights-motivation-emotions-eeg/)
4/3/2017 6:07 PM
sydneyneurotechx :
They also have a great SDK. I would just love to see more open source projects with it (There are a few, but I would love to see more)
4/3/2017 6:16 PM
maxim :
Yeah, their SDK was a big motivation to buy the headset. What kind of OSS projects with Muse would you want to see? I'm interested in doing something with it now, but having a better understanding of what's possible will be great
4/3/2017 6:17 PM
maxim :
Also, we are kind of stuck with writing smartphone apps for now due to the SDK mostly working on iOS and Android.
4/3/2017 6:22 PM
maxim :
Thanks for the frontal asymmetry link! On challenge here is that there seems to be a lack of publicly available Muse datasets, so the first thing would be to make an app to record and store them for later analysis
4/3/2017 6:39 PM
sydneyneurotechx :
The eeg 101 app that dano and team have created, allows for storing of datasets. You can take a look at at. What I'm personally motivated to see is a framework/sandbox to use the muse with VR. It's a pretty good form factor for it (Not perfect though. VR glasses and the Muse fit well on my head. It may not on others). I'd also love to see some open source algorithms for different mental states.
4/3/2017 6:41 PM
maxim :
Yeah, I saw that eeg-101 is starting to store data now, although I'm interested in storing raw data and not a processed CSV, don't want to potentially miss out on some signal (although that might be my lack of BCI knowledge). Storing data doesn't seem that hard anyway, Muse's own demo app does it, so it is literally just extending it to allow you to save it in different files.
4/3/2017 6:42 PM
sydneyneurotechx :
I believe they store all date in the CSV. It's just a different filetype. dano Can answer this better than I can though.
4/3/2017 6:42 PM
maxim :
VR is potentially an interesting application, but even harder to find people to do it it seems. I don't have a VR headset, so there is no way for me to debug it and I bet an intersection of people that have both VR and Muse is pretty small. Mental state detection is something that seems more interesting personally.
4/3/2017 6:44 PM
maxim :
That's fine, storing data is one of the first trivial steps, so no harm in redoing it to get used to working with Muse SDK. I'm thinking of also storing a bunch of stuff that's also available from a general phone (mic for noise level and light sensor maybe?) to have more variables to correlate with later on
4/3/2017 6:57 PM
sydneyneurotechx :
cool! The eeg-101 tutorial is open source so you can take a look at their project and see how they do it if you get stuck.
4/3/2017 7:05 PM
dano :
Yea, recording Muse data isn't that hard, but do take a look at the EEG 101 code if you get stuck. I found that the Java/Android stuff was the harder part.
Also, maxim, as you may know we're working on building an open source Muse dataset storage and analysis pipeline with Cloudbrain and we could definitely use some help on that front. Let's talk!
4/4/2017 8:48 AM
mertnesvat :
Cool post 🙂 I would like to add some scientific papers to idea 1b)
http://wearcam.org/inventometer real light bulb like in cartoons 😄
http://feeler.evadurall.com|feeler.evadurall.com attention and cognitive load tracker which uses neuroSKY
4/4/2017 4:22 PM
tenoke :
another common example for 2) - games. Also I'd be more explicit in what is possible today (and in general with eeg) - up/down yes (or binary options in general) but not much more (e.g. full control of a character).
i'd also add more examples (with years next to them) and more details on the limitations of eeg (so newral hackers can figure out if it's likely that their current idea is not possible without a more invasive procedure like an implant)
4/5/2017 2:00 AM
stephen :
Thanks mertnesvat
Hey you other people reading this post, what else do you agree or disagree with? 🙂
4/5/2017 7:03 AM
frederictaieb :
Hi All, I've just received my 2014 Muse. I am trying to make it work with a Raspberry Pi. Unfortunately, I have understood that Interaxon didn't develop their SDK for ARM Systems. I was wondering if some of you were using RPi+Muse and if anybody could give me some advice to start. Thank you all.
4/5/2017 10:52 AM
yannick :
Hey frederictaieb, yes, atom2626 & ron.brash developed IntelliPi an EEG library for the Pi using Muse headset.
You might want to look at:
4/5/2017 12:44 PM
omer :
Yes, Thank you. I guess I agree with the general impression that things are more complicated than they seem, in comparison to software neurohacking requires more knowledge and understanding, and the quick and dirty startup practice people are used to may not work.
More specifically, regarding 1see the attempts to incorporate sensors in already wearable tech such as glasses and jewlery (as the bracelet went into a watch - same will follow here). regarding 2 this the same, while you already wear it... anyway I haven't seen any convincing products so far that are not based on muscles perhaps the new gtec initiative with als wil prove me wrong.
3 - this far more tricky, I guess most the scitifc community still truely do not know do tdcs/tacs work, and there are few papers you can trust. It seems that every good one has a startup trying to implement it to the market. I think people try to go to market too fast, if every entrepreneur would try first to replicate the findings he is basing on things will look different.
4/5/2017 12:52 PM
benjamindeleener :
frederictaieb PolyCortex also developed a Python library receiving data from the Muse (https://github.com/PolyCortex/pyMuse) but it still needs Muse connectivity and we did not tested it on the RPi
4/5/2017 4:35 PM
frederictaieb :
Thank you yannick benjamindeleener . I guess I will start with intelliPi then, even if if would much easier for me to use a python lib.
4/5/2017 4:36 PM
frederictaieb :
yannick, do you know if IntelliPi is working correctly on a Pi3?
4/5/2017 5:55 PM
sydneyneurotechx :
ron.brash ^.
4/6/2017 6:25 PM
kowalej :
Hey all, not sure if you've heard of the Brain Scope EEG but a new study out of John Hopkins has shown this handheld device can identify brain bleeding in people with head injuries at a 97% accuracy. The idea is this would be a low cost portable pre scan tool. Could be an interesting device for us hobbyists to get our hands on. Here's the article: http://www.hopkinsmedicine.org/news/media/releases/quickly_assessing_brain_bleeding_in_head_injuries_using_new_device
4/6/2017 6:56 PM
sydneyneurotechx :
Yea, met a former engineer at Brainscope. I like their approach to using eeg for concussion detection.
4/6/2017 7:00 PM
sydneyneurotechx :
Can't find any tech specs on the AHEAD 300. Any ideas?
4/6/2017 8:23 PM
kowalej :
I reached out to the company and they will get back to me shortly with information I should be able to share.
4/6/2017 8:40 PM
sydneyneurotechx :
sweet! Thanks!
4/6/2017 9:41 PM
sydneyneurotechx :
Anyone here know of this company? They are based out of Japan http://www.araya.org
4/7/2017 12:31 AM
breeves :
Anyone have a good papers or readings on noninvasive ways to map out neural activity i.e., not implanting electrodes
4/7/2017 12:32 AM
breeves :
Basically a super hard signal processing problem
4/7/2017 12:32 AM
breeves :
Any interesting reads on optical or electromagnetic approaches are much appreciated
4/7/2017 12:49 AM
yannick :
You mean the Inverse Problem? (Source Localization)
4/7/2017 1:23 PM
breeves :
yannick: yes but looking into alternatives to the Elon / Bryan Johnson approach of implanting hardware
4/8/2017 12:45 AM
stephen :
Also interested in this. What approaches or projects do you think are interesting so far? How would you go about searching the field to find others?
4/8/2017 4:45 PM
prefrontalvortex :
breeves phase slope index is a pretty new idea and it's pretty exciting imho https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5171906/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2996143/
4/9/2017 5:17 AM
nirmeshg :
4/9/2017 5:54 AM
nirmeshg :
Still a theory until additional data is out by the Gaia project in the next 5 years. Still an interesting article.
4/9/2017 2:54 PM
francescacoo :
Hi everybody, I was wondering if anybody knows of some event/summer school/something where to learn something more on neuroscience or anything related to BCI. I applied for a summer school in machine learning in Lisbon, waiting to see if they accept me, so I was thinking to something similar. I have a degree in IT and currently doing a master in Software Engineering, so unfortunately something where the requirements are not too high (btw I am in Europe)
4/9/2017 3:45 PM
andreacoravos :
francescacoo harvard's intro to neuroscience mooc is supposed to be good https://www.mcb80x.org/faq
4/9/2017 5:20 PM
pierre :
There's a list of Summer schools here: https://github.com/PhABC/neuroSummerSchools/blob/master/README.md
4/10/2017 6:50 AM
frederictaieb :
Hi all, for fun, I would like to make a project that read the MUSE (Iphone or RPi), detect if I am calm and if I am calm it communicate to a raspberry pi to trigger a LED. Do you know if there are some websites where I can find this kind of information. Thank you very much for your help. Have a great day.
4/10/2017 3:47 PM
dano :
Hey Frederic,
If IntelliPi hasn't worked out, I'd suggest an alternative approach of reading the Muse data from an iPhone app and then sending data to the Raspberry PI through websockets. Here's a decent tutorial on how to set up a Node.Js server on an arduino that receives LED controls over websocket
4/10/2017 7:07 PM
frederictaieb :
Thank you Dano!
4/10/2017 7:07 PM
frederictaieb :
I will take a look 🙂
4/12/2017 4:27 AM
sydneyneurotechx :
Using Human Brain Activity to Guide Machine Learning https://arxiv.org/pdf/1703.05463.pdf
4/12/2017 8:17 PM
andreacoravos :
We've been drafting more stories for the NeuroTechX medium publication. Just published a post on the BRAIN Initiative: https://medium.com/neurotechx/the-brain-initiative-the-race-to-understand-the-human-brain-c2fddfea271e
4/13/2017 7:32 PM
kowalej :
Didn't get much info on the Ahead other than it costs $10k+ for the reader and each single use headband with sensors costs $300. Also there is no open way to access the data other than with their reader other than reverse engineering it.
4/13/2017 8:34 PM
sydneyneurotechx :
No Surprise. Funny that they made it single use headband.
4/15/2017 10:59 PM
sydneyneurotechx :
These new status emotes are awesome. If you have suggestions of others to add, let me know.
4/17/2017 3:22 AM
yannick :
Hey dmitryneuro, welcome on board! Please introduce yourself in <#C08QEC5H7|introductions> as many people might be interested in discussing hardware with you in the <#C08T2SENQ|devices> channel 🙂
4/18/2017 8:41 PM
sc :
4/18/2017 9:30 PM
yannick :
I used to be fast on BCI/Neurotech news, but now sc you�re beating me every time 😞 Gotta spend less time on that PhD&
4/18/2017 9:31 PM
yannick :
Interesting! PR & Marketing stunt or real science and/or commitment we shall see.
4/19/2017 5:20 AM
sc :
Working hard on my honorary PhD instead
4/19/2017 7:06 AM
physionikhil :
Everyone of you motivate me to pursue a PhD in neurosciences
4/19/2017 7:57 AM
alexandre.barachant :
Yes, eager to see what they want to do. I will also be interested to see who they managed to recruit
4/19/2017 8:03 AM
alexandre.barachant :
They still have two identical job opening for this
4/19/2017 11:02 AM
benjamindeleener :
#phdislife
4/19/2017 2:53 PM
wanfuse123 :
hi! I am wondering if anyone has seen this https://medicalxpress.com/news/2017-03-brain-tranquility.html and wondering if neural stimulation might apply to this meditation center of the brain...thoughts anyone? Sorry for the double post but I posted it in the wrong place 😉
4/19/2017 6:31 PM
bechir :
More info on Facebook�s BCI: �a team of 60 engineers working on building a brain-computer interface that will let you type with just your mind without invasive implants�
https://techcrunch.com/2017/04/19/facebook-brain-interface/
4/19/2017 6:38 PM
graeme :
Did facebook just announce David Eagleman's sensory substitution vest and a P300 speller?
4/19/2017 6:51 PM
joeyo :
fNIRS but yes
4/19/2017 6:51 PM
joeyo :
the not-based-on-hemodynamic response part is interesting
4/19/2017 6:52 PM
joeyo :
curious to see if they can really do single-trial decoding
4/19/2017 6:55 PM
yannick :
dvidsilva is there live, no? (regarding graeme�s question ^)
4/19/2017 7:00 PM
graeme :
joeyo fast optical signal?
4/19/2017 7:01 PM
joeyo :
http://www.theverge.com/2017/4/19/15360798/facebook-brain-computer-interface-ai-ar-f8-2017
4/19/2017 7:02 PM
joeyo :
�Implanted electrodes simply won�t scale,� she says. �We think optical imaging is the best place to start.� Using neural imaging may be the only non-invasive approach to transmitting neural activity into inputs for electronic devices, Dugan says. This could be some type of cap you wear on your head, but Dugan adds that this technology does not exist yet and would have to be developed over the course of many years.
4/19/2017 7:06 PM
sydneyneurotechx :
Are you being sarcastic or are they actually building pre-established tech?
4/19/2017 7:23 PM
yannick :
================================================
4/19/2017 7:25 PM
yannick :
========================================
Here is the Facebook Brain-Computer Interface talk.
Head of Building 8 - Starting @ -24:10
Good Stuff - Starting @ -18:38
https://www.facebook.com/FacebookforDevelopers/videos/10154616829398553/
========================================
4/19/2017 7:36 PM
yannick :
joeyo: the �not-based-on-hemodynamic response part� is at: -12:12
https://www.facebook.com/FacebookforDevelopers/videos/10154616829398553/
�> measuring the change in "the neurons as they suck in sodium and spit out potassium"
4/19/2017 8:45 PM
aj :
are they talking like fnirs tech?
4/19/2017 8:51 PM
yannick :
Same idea, you send a light and you record it as it goes out and you do inference on what happened in between.
fNIRS (given the wavelength) is useful for blood flow.
FB is aiming at measuring the changes in neuron properties based on sodium/potassium changes, which would be a more direct measure of �brain activity� and not �blood activity�. But I don�t know much about how the light absorption properties of a neuron change as its electrical potential changes.
4/19/2017 8:54 PM
yrenard :
And then after that, they are aiming at capturing the semantic to let people who speak different languages be able to communicate, so you can call your mom (which btw may speak the same language as you do)
4/19/2017 9:37 PM
graeme :
That's fast optical signal, or optical evoked potentials
4/19/2017 9:48 PM
sydneyneurotechx :
Are they serious??? 100 WPM??
4/19/2017 9:48 PM
yrenard :
🙂
4/19/2017 9:49 PM
yrenard :
because of semantic dude
4/19/2017 9:49 PM
sydneyneurotechx :
Is there any papers out there they are basing this on?
4/19/2017 9:56 PM
sydneyneurotechx :
I see what you mean now. Yea it sounds like they just took the concept of Neurosensory but shrunk it down
4/19/2017 10:03 PM
sydneyneurotechx :
Only good reference I can find on Balistic Optical Imaging is from this book. https://books.google.ca/books?id=EJeQ0hAB76gC&pg=PR3&redir_esc=y#v=onepage&q&f=false
4/19/2017 10:04 PM
sydneyneurotechx :
Anyone find anything better?
4/19/2017 10:16 PM
graeme :
4/19/2017 10:16 PM
graeme :
if you take Bob Knight's work on speech imagery:
4/19/2017 10:17 PM
graeme :
http://knightlab.berkeley.edu/statics/publications/2016/10/03/srep25803.pdf
4/19/2017 10:18 PM
graeme :
and you figure out a way to improve on that, and then also make this quasi-ballistic fast optical response imaging work with high spatial resolution and lots of sensors...
4/19/2017 10:19 PM
graeme :
...it's kind of not crazy
4/19/2017 10:19 PM
graeme :
just a lot of work to make it happen
4/19/2017 10:19 PM
graeme :
if it can be done
4/19/2017 10:23 PM
yrenard :
not because the measurement may (or may not) make sense that the overall thing does, and given all of the second half of the talk, I bet that they have been terribly misleaded on the problem they are trying to solve (at the best)
4/19/2017 10:54 PM
sydneyneurotechx :
Awesome References. Added them to the awesome-bci list
4/19/2017 10:54 PM
sydneyneurotechx :
Added the pdf and books to the awesome bci list if people want to look into it later on https://github.com/NeuroTechX/awesome-bci
4/19/2017 11:35 PM
joeyo :
http://www.sciencedirect.com/science/article/pii/S1364661300017010
4/20/2017 12:39 AM
sydneyneurotechx :
For all those who would be interested in talking at the DEF CON Biohacking village, their call for papers is closing soon. Click below for more details
https://www.defconbiohackingvillage.org/
4/20/2017 6:17 AM
sc :
You're all way ahead of me today, but this link features videos http://mashable.com/2017/04/19/facebook-brain-interface/
4/20/2017 12:41 PM
alexandre.barachant :
Looks like they are going for Fast Optical imaging .... good luck to them
4/20/2017 12:55 PM
alexandre.barachant :
The part about quasi-ballistic photon is totally missleading. FOS is an almost 2 decade old technology. they have to find something that will look like they are inovating.
this does not solve any of the problem this technology has (1- we don't know what we measure 2- the signal to noise ratio is worse than any other non-invasive tech, including EEG)
4/20/2017 1:01 PM
yannick :
When you say the signal to noise ratio is worse, do you think its a mathematical problem (optimization / stochastic problem) or hardware related or physics related (i.e. there is just no perfect wavelength for what we�re targeting)?
4/20/2017 1:14 PM
alexandre.barachant :
The few papers showing that we can record fast brain activity with FOS are based on ERP type of analysis. It require average of hundreds of stimulation to show a response. Thats a good starting point, but for a BCI type of application, you need a clear signal on the single trial level. What i'm saying is that okay, the spatial resolution is better, but you need to improve the tech by a factor 100 to actually get something that will allow operation in real time.
Using quasi-balistic light is just an oversimplistic response to a very tricky problem. This will just allow to focus a bit better the beam (increasing the spatial resolution) but does not solve the SNR problem.
The way i understand thing is that FOS works by measuring change in optical properties of neurons when they are activated. So you have to send a beam of light in the right place, it get diffracted by the population of neurons firing and somehow has to come back to the sensor. there is a lot of loss in the process. IMO limitation are physical and hardware related.
4/20/2017 1:22 PM
alexandre.barachant :
Anyway, i hope they succeed, that will be really great.
4/20/2017 1:34 PM
oriansha :
hi
4/20/2017 1:51 PM
sydneyneurotechx :
hi oriansha ! If you are new to NeuroTechX, please introduce yourself in the <#C08QEC5H7|introductions> channel 🙂
4/20/2017 3:25 PM
dano :
what else would you do if the people that hired you had no idea about your field but wanted results at the end of the quarter? 😛
4/20/2017 5:26 PM
graeme :
this is a great discussion alexandre.barachant yannick
4/20/2017 5:26 PM
graeme :
I love this slack channel!
4/20/2017 6:31 PM
joeyo :
Possibly relevant: https://www.youtube.com/watch?v=WsIWnMazQdc
4/20/2017 6:35 PM
yannick :
alexandre.barachant joeyo we should organize a AMA with Mary Lou Jepsen. 🙂
4/20/2017 7:12 PM
graeme :
i mean, come on... optogenetics + fast optical fNIRS to control neuroplasticity non-invasively... this has never even been done in mice.
4/20/2017 7:15 PM
graeme :
it's statements like those that make it hard for the neuroscience community to take the BCI community seriously
4/20/2017 7:20 PM
dano :
I'm not convinced human optogenetics is ever a thing that should be done. DREADDs (which are pharmacologically vs optically activated) maybe...
4/20/2017 7:28 PM
bryan_j :
ofMRI all the way baby
4/20/2017 7:29 PM
bryan_j :
4/20/2017 11:45 PM
shivanisheopory :
4/21/2017 5:31 AM
aj :
shivanisheopory that article is so damn good!! I'm halfway through and am for sure gonna read it all again
4/21/2017 11:48 PM
change :
If anyone�s interested in the field of psychedelic neuroscience - the conference is being streamed live: http://psychedelicscience.org/live
4/22/2017 6:08 PM
okbalefthanded :
reaching a spelling rate of 100 word per minute spelling using this technology seems hard to achieve in 3 years, the fastest speller demonstrated so far using EEG and SSVEP showed a 60 character per minute rate.
http://www.pnas.org/content/112/44/E6058.abstract
4/23/2017 12:56 AM
vitruvia :
Hey guys, is there any way to input into a PC with eeg? As if it were a mouse or keyboar I mean. I read that some group at MIT was working on this but wanted to know if this was actually a thing
4/23/2017 1:05 AM
sydneyneurotechx :
Hey Vitruvia, as in you want to control a mouse and keyboard with an eeg?
4/23/2017 1:05 AM
sydneyneurotechx :
If that is the case there are different options that exist at different pricing points
4/23/2017 1:06 AM
sydneyneurotechx :
The answer will also depend on your technical experience
4/23/2017 1:24 AM
vitruvia :
I haven't got any actually. I just know some very basic python. I'm just wondering what is out there so I can choose what to study.
4/23/2017 1:24 AM
vitruvia :
and yes I meant something similar to controlling a M&K with eeg
4/23/2017 2:59 AM
vitruvia :
sydneyneurotechx sorry for the delay in my response 😃
4/24/2017 2:18 AM
sydneyneurotechx :
If you are ok in spending a bit of money, you can go with the emotiv epoc option. The epoc will set you back about 800+
4/24/2017 2:18 AM
sydneyneurotechx :
and can go along with this software
4/24/2017 2:18 AM
sydneyneurotechx :
aj may have an option for you with the openbci, but not totally sure
4/24/2017 10:48 AM
jmhorschig :
alex, this is correct. We think (i.e. are sure) that the FOS does not work with CW fNIRS devices (the most commonly used devices), so I am fairly sure that FB will not pursue this path or fail miserably.
4/24/2017 11:35 AM
vitruvia :
oh. I don't have that much money right now. And also I'm more interested in learning how software of this type is built then on getting it ready to use
4/24/2017 12:44 PM
alexandre.barachant :
jmhorschig Very interesting. Did you tried to replicates results from the few FOS papers there exist ?
If it's not FOS they are pursuing, then What ? speech decoding (at least at 100 WPM) can not be achieved just with the BOLD response ...
4/24/2017 12:45 PM
alexandre.barachant :
actually, it's clear from one of their slide that they are not going for the Bold response
4/24/2017 1:38 PM
jmhorschig :
If they really want to go for 100words/minutes, they have to go for FOS/EROS, hemodynamic responses would be too sluggish and convoluted- I just think it does not work. We tried to replicate such work (I work for Artinis, a NIRS company), and my boss and his former colleagues from academy tried so as well. None had success (and that includes top NIRS researchers which are still active in the field). It happens to be that just approximately one group is recording FOS and publishing about this. Some more groups are coming up now, but it is really not something that you just do along the way. Also, the current record with BCI is around 120bits/minutes (EEG+SSVEP). That is extremely far off from 100 words/minutes.
I think what they might be hoping for is that quasi-balistics photon theory, maybe indeed in combination with what Mary-Lou Jespen is developing, can be combined and constitute a paradigm shift. I am very critical on this though.
4/24/2017 4:13 PM
alexandre.barachant :
Thanks for the detailed answer. FOS was bugging me a lot because as you said, only one group is publishing about it. it's been more than a decade and i did not see a clear replication so far. You are confirming my hunch.
The work of ary-Lou Jepsen seems very interesting and promizing, but to me it's more about anatomy than function. Everything based on BOLD response will never provide the temporal resolution to capture enough information to decode speech (i might be wrong ...).
IMO, for real time speech decoding, you need a high spatial resolution and be able to tap into the gamma activity. The best way to achieve this non invasive is with MEG ... perhaps in 10 years.
4/24/2017 5:47 PM
sydneyneurotechx :
So perhaps working with an eeg database first?
4/24/2017 5:49 PM
sydneyneurotechx :
I guess an approach you could use to bring you up to speed, is first trying to do some analysis on different brain activities (ie eeg paradigms), learn some of the machine learning methods to classify mental states, and then building a "translation" layer which takes different mentals stales and ties it to a computer input.
4/24/2017 5:49 PM
sydneyneurotechx :
Not sure if what I wrote is clear or not. Does it make sense?
4/24/2017 7:58 PM
stephen :
Who knows people working on automatic human behavioral analysis, such as interpretation of body movements, eye tracking, HRV, facial expressions? Anyone working on hardware systems that integrate some of those data sources?
4/24/2017 7:59 PM
amilenkovic :
Hi Vitruvia, maybe I can provide some input as well. You can use an EEG as a method of providing inputs to a device. The way it typically works is that you use an EEG device to find the pattern you are looking for. That becomes your input. So when someone thinks of the color blue, you will move the cursor up. When they think of green, you will move the cursor down. You record what their brain looks when thinking of Green. Then you record what their brain looks like thinking of Blue. Then you write a program that says "When the brain waves look like what I saw when the subject was thinking green, move the cursor up. When it looks like blue, move the cursor down." The technology is available today and works reasonably well. The limitation is how many things you can reliably distinguish and match on.
4/24/2017 8:20 PM
sydneyneurotechx :
I'd suggest talking to the iMotions people. They have a multi-modal bio input system for that type of analysis
4/24/2017 8:21 PM
sydneyneurotechx :
They'd also know people who are
4/24/2017 8:21 PM
sydneyneurotechx :
I also believe that watson has an interest in this
4/24/2017 9:52 PM
watson :
My interest is in inferring psychological/emotional state and I�ve been focused on physiological signatures, but also informed by things like accelerometer, to give context& stephen when you say �automatic human behavioral analysis� what do you actually have in mind? Do you mean walking vs running vs sitting? Attending vs not attending? Gesturing? Emotion? Whats the question you�re interested in?
4/24/2017 9:54 PM
watson :
also I started the channel <#C4NA4CTK3|autonomic-system> for discussing signals from the body/periphery, which could include things like sweat and muscle tension. join in if thats of interest
4/24/2017 10:01 PM
stephen :
watson Yes, specifically things related to monitoring cognitive states.
4/24/2017 10:20 PM
watson :
you probably know of Affectiva already, a spinoff from the MIT Media Lab: http://www.affectiva.com/
4/24/2017 10:22 PM
vitruvia :
sydneyneurotechx yes it makes sense. I'm a beginner so this is more of a long term goal for the year. Where could I learn about EEG databases and about machine learning method specific to this task?
4/24/2017 10:23 PM
vitruvia :
amilenkovic Thanks! That is how I thought it might work, glad to see I wasn't so off the mark. What you said is also in line with what sydneyneurotechx suggested, so I guess learning how to analyze and translate brain activity would be a good start
4/24/2017 10:23 PM
vitruvia :
I still need to learn machine learning, though. Like I said, I'm a big beginner. Hopefully I'll be able to study it by the middle of this year
4/24/2017 10:25 PM
amilenkovic :
You can match on patterns without going into ml. Netflix used an accelerometer in an eeg, and most demos use blinks or jaw clenched. Many vendors already have a blink algorithm, you can use that, and have the blink do something in an app
4/24/2017 10:25 PM
amilenkovic :
So it just depends on how fancy of a setup :)
4/24/2017 10:30 PM
sydneyneurotechx :
First off take a look at the awesome-bci list if you haven't https://github.com/NeuroTechX/awesome-bci
In terms of databases, You can look at SCCN's list: https://sccn.ucsd.edu/~arno/fam2data/publicly_available_EEG_data.html
You can analyze that data in Matlab or Python. If you want to try Python, I'd suggest using the MNE library. https://www.martinos.org/mne/stable/index.html
The thing you need to understand before moving forward is that working with eeg and learning to code is something which takes some time to do it properly. Give yourself enough time and be patient. You may need to build some foundational knowledge before you get into the advanced stuff. Depending on your speed, it may take some time (This is part of the reason why many people get frustrated and stop) I'd suggest finding a friend to do this with (or perhaps someone in this slack). In some cases, this can be a semester long project, so be patient and have fun with it 🙂
4/24/2017 11:59 PM
vitruvia :
Alright. Thank you both
4/25/2017 12:00 AM
vitruvia :
sydneyneurotechx Don't worry, I see this as a long-term project anyways. Just needed a starting point. Thanks again! =]
4/25/2017 2:04 AM
david :
Also, she has referenced this paper when she talks about how the study of diffuse optical tomography was at a standstill for several years until this paper was released. She mentions that this paper helped diffuse optical tomography go from an intractable math problem, to reality - http://www.math.uni-frankfurt.de/~harrach/publications/uniqueness.pdf
4/25/2017 2:30 AM
sydneyneurotechx :
great find david . Will add it to awesome bci list
4/25/2017 7:25 AM
jmhorschig :
david: actually, she wants to revolutionize the NIRS scene by using hologram techniques appyling a technique called phase conjugation (see https://www.youtube.com/watch?v=gAy39ErqV34 for a demonstration). Also check out her AMA on quora, e.g. https://www.quora.com/What-are-the-key-advantages-of-the-holographic-method-used-in-OpenWater-versus-diffuse-optical-tomography
4/25/2017 7:28 AM
jmhorschig :
anyway, thx for the book and the paper, sounds definitely like another part necessary to solve the riddle on what she wants to pursue (but then seen from the algo/SW side) ;)
4/25/2017 11:32 AM
david :
jmhorschig thanks for this as well! so a combination of diffuse optical tomography + phase conjugation to be able to undo the effects of the light scattering through the brain? -if I'm understanding correctly on the last part
4/25/2017 8:58 PM
th :
Hello
4/25/2017 9:00 PM
th :
Any recommended papers on decoding brain signals?
4/25/2017 9:00 PM
yannick :
Hi th did you check the awesome BCI list:
https://github.com/NeuroTechX/awesome-bci
4/25/2017 9:02 PM
th :
Thanks for that will tackle it
4/25/2017 9:04 PM
sydneyneurotechx :
The code's open source so go for it. https://github.com/NeuroTechX/NeuroBotX
4/25/2017 9:21 PM
bryan_j :
yannick: dano can you think of anyone from our teams/community to assign?
4/25/2017 9:28 PM
dano :
We've definitely met somebody with language processing experience. Can't remember where...
4/26/2017 8:00 AM
jmhorschig :
yes, exactly, I think that is their ultimate goal. Apparently they also applied for patents, but we'll just see them when they are approved.
4/26/2017 8:07 AM
jmhorschig :
It's very hard with MEG (colleagues of mine did that, see http://www.mpi.nl/departments/neurobiology-of-language/projects/mous-mother-of-all-unification-studies). Not sure if gamma is sufficient. I think what FB wants is to use NIRS and hope for a combination of "electrophysiological" signals using FOS and hemodynamic responses, thereby getting the best of both worlds. What might work is some combination of an error-signal and auto-write based on previous conversations and the current conversation. So, something similar to this: http://www.theverge.com/2016/1/21/10805398/friends-neural-network-scripts and then detecting if the predicted/suggested phrasing is what the user intends to say. When writing about this, yes, I think this is what they plan to do. It's the only thing that is probably possible and fits with the knowledge and experience FB already has. And it might work up to a certain extent. Users will probably be like "nah, I didn't want to write exactly that, but something along these lines, so ok. Send!". But that'll be more the deep NN doing the work being individualized by having analyzed previous conversations. What do you think?
4/26/2017 9:33 PM
hectordomorozco :
:ntxblue: 💖 :ntxblue:
4/27/2017 5:53 AM
w :
Is anyone from CeReB on this slack?
4/27/2017 1:09 PM
yannick :
Yes gena_cereb ^
4/27/2017 1:15 PM
bryan_j :
yannick: you beat me to it! 🙂
4/27/2017 1:16 PM
bryan_j :
Amazing! Good group photo :) hectordomorozco NTTO is here to support and collaborate!
4/27/2017 1:17 PM
yannick :
I let you the whole night to reply& You didn�t& What were you doing? 😛
4/27/2017 1:17 PM
yannick :
Less Sleep, More Slack.
4/27/2017 2:23 PM
hectordomorozco :
Thanks bryan_j :D
4/27/2017 2:58 PM
bryan_j :
yannick: hahaha my attention was toward other tasks, unfortunately. I had a night of less sleep, less Slack :white_frowning_face:
4/27/2017 3:53 PM
w :
amilenkovic
4/27/2017 3:58 PM
amilenkovic :
ah cool, thanks 🙂
4/27/2017 4:07 PM
graeme :
w I think I'm also in CeReB
4/27/2017 4:50 PM
chrisabbott :
Hey guys. I'm currently writing a few python scripts to interface with Muse (2016 model) using GATT on Linux. I have the raw attribute output from the device, but I've been unable to find any information online about which data stream each UUID/attribute corresponds to (IE; battery life attribute vs sensor attribute). Does anybody know where I can find a list of the device attributes and UUIDs for Muse?
4/27/2017 6:31 PM
pierre :
chrisabbott I don't know the answer to your exact question, but alexandre.barachant wrote a python script to interface with muse via GATT
4/27/2017 6:31 PM
pierre :
It is here
4/27/2017 6:31 PM
pierre :
4/27/2017 9:29 PM
chrisabbott :
pierre Ah this is perfect. Thanks!
5/1/2017 7:51 PM
sydneyneurotechx :
<!channel> I would to announce NeuroTechX's new initiative, the "Mother of all BCI Benchmark" project.
The Goal of this project is to build a comprehensive benchmark of popular BCI algorithms applied on an extensive list of freely available EEG dataset. The code will be made available on github, serving as a reference point for the future algorithmic developments. Algorithms can be ranked and promoted on a website, providing a clear picture of the different solutions available in the field.
Reproducible Research in BCI has a long way to go. While many BCI datasets are made freely available, researchers do not publish code, and reproducing results required to benchmark new algorithms turns out to be more tricky than it should be. Performances can be significantly impacted by parameters of the preprocessing steps, toolboxes used and implementation �tricks� that are almost never reported in the literature. As a results, there is no comprehensive benchmark of BCI algorithm, and newcomers are spending a tremendous amount of time browsing literature to find out what algorithm works best and on which dataset.
alexandre.barachant will lead the initiative and I will provide logistic support to push it forward. This will be mostly a coding project (It will be open source, so anyone can contribute), however people with a design background are invited as well.
If you are interested in participating, please fill out the doodle below. We will arrange one or two calls based on everyones availability.
https://doodle.com/poll/4ngxy9h8qpcmkhpb
More details on the project can be found here:
https://docs.google.com/document/d/18nU07Qci_VDbzZCLveDQqPgqRODt7HCNBfKeNPVwMQY/edit#
5/1/2017 8:33 PM
nicolas :
sydneyneurotechx: Very cool project ! 👍
5/2/2017 12:10 AM
marion :
sydneyneurotechx awesome. I'll be on the call. We have created a benchmark at numenta for online anomaly detection, to allow people to compare different algos on a variety of datasets :https://github.com/numenta/NAB. Maybe interesting. There are many other benchmarks in ML. Another one that I found interesting is the UCR timeseries dataset (to benchmark classification algos on temporal data)
5/2/2017 12:10 AM
marion :
5/2/2017 1:11 AM
pat :
Just to check I understand: this would be similar to creating standard datasets like mnist/cifar, and collecting in one place to make it easy to run against and benchmark the results?
5/2/2017 9:06 AM
alexandre.barachant :
pat yes, kind of. we don't have a cifar/minst dataset, but there is many small dataset available. This project is about wrapping them around the same format so we can have a more robust benchmark and get a better picture of what's works and in which context
5/2/2017 9:31 AM
rohit :
http://sapienlabs.co/building-global-collaborations/ Anyone here involved in this project? Would like to have a quick chat?
5/2/2017 12:09 PM
yannick :
Hey rohit: tara is the founder of the initiative.
We had a good chat 2-3 months ago, here is a small piece about NTX on SapienLabs� blog: http://sapienlabs.co/neurotechx/
5/2/2017 5:31 PM
rohit :
Thanks yannick. I am aware of the NTX article, just came across their new article and wanted to explore a bit.
5/3/2017 7:34 PM
yannick :
Not related, but there seems to be something going on with Google Docs, beware.
I received couple of invite to share G-Doc from different people in the last 30 mins, which fortunately I didn�t open, then I Googled it, and I�m definitely not alone&
https://www.reddit.com/r/sysadmin/comments/692cno/sudden_google_docs_spam/
https://motherboard.vice.com/en_us/article/massive-gmail-google-doc-phishing-email
5/3/2017 7:35 PM
bryan_j :
Thanks for sharing yannick
5/5/2017 1:21 PM
etakla :
For anyone who clicked the fake link: http://www.cbc.ca/news/technology/google-docs-gmail-phishing-attack-1.4099238?cmp=rss
5/10/2017 10:25 PM
stephen :
Poll: What are your favorite works of science fiction that demonstrate brain tech or human intelligence enhancement?
Initial examples:
- The Nexus Trilogy, by Ramez Naam
- The Aurora Rhapsody - G.S. Jennsen (�a combo of space exploration, HI, and AI.� - Tom)
- Flowers for Algernon - Daniel Keyes (an old classic)
5/10/2017 10:44 PM
yannick :
stephen did you fail the Poll command? We can�t answer& Unless you want to manually count them 😛
5/10/2017 10:45 PM
yannick :
The Nexus Trilogy +1 (Ramez Naam)
Neuromancer +1 (William Gibson)
5/11/2017 9:30 AM
catpitch :
Flowers for Algernon
5/12/2017 12:50 AM
jfrayshe :
Ghost in the Machine is more cognitive philosophy but I've always enjoyed it as a philosophical input to the subject.
5/12/2017 6:34 AM
pierre :
The movie "Strange Days" was pretty cool
5/12/2017 12:42 PM
cbattista :
iain m. banks' 'culture' novels with the neural lace is pretty neat
5/12/2017 12:44 PM
cbattista :
5/12/2017 1:43 PM
yannick :
davidevaleriani: �But what exactly is Neuralink? If you have time to read a brilliant 36,400-word explainer by genius Tim Urban, then you can do so here. If you don�t, Davide Valeriani has done an excellent summary right here on The Conversation. However, to borrow a few of Urban�s words, NeuraLink is a �wizard hat for your brain�.�
5/12/2017 5:03 PM
watson :
stephen: reaching back in time, there�re a few examples of memory implant stories. Gibson�s Johnny Mnemonic, with (pretty bad) film adaptation of the same name. and even further back, this story by Philip K Dick, which later became Total Recall: https://en.wikipedia.org/wiki/We_Can_Remember_It_for_You_Wholesale
5/12/2017 5:04 PM
watson :
If you want to reach even further back in time there is PDKs short story �The Unreconstructed M� from the 1950s, where there�s a door that can be unlocked by brain activity: http://www.technovelgy.com/ct/content.asp?Bnum=2486
5/12/2017 5:08 PM
watson :
Peruse this page for 21 search results for �neur� and 81 for �brain�, some dating back to the 1930s! http://www.technovelgy.com/ct/Science_List_Detail.asp?BT=Medical
5/12/2017 5:13 PM
stephen :
Awesome suggestions, thank you watson! Would love to keep building up a library of these neurotech stories!
5/12/2017 6:02 PM
watson :
Oh wow, forget the 1930s& This one is from 1705! �This Machine that I am speaking of, contains a multitude of strange Springs and Screws, and a Man that puts himself into it, is very insensibly carried into vast Speculations, Reflexions, and regular Debates with himself: They have a very hard Name for it in those Parts; but if I were to give it an English Name, it should be call�d, The Cogitator, or the Chair of Reflection.� http://www.technovelgy.com/ct/content.asp?Bnum=2697
5/15/2017 2:32 PM
dano :
The Truth of Fact, the Truth of Feeling by Ted Chiang
http://subterraneanpress.com/magazine/fall_2013/the_truth_of_fact_the_truth_of_feeling_by_ted_chiang
5/15/2017 10:35 PM
francescacoo :
What do you think of this? http://summerschool.ssa.org.ua/program/computational-neuroscience
I am studying for a master in software engineering and no academic background in neuroscience. While I will use online resource a to learn (and neurotechedu!) I would love to do some course/summer school.. this is the only one I found so far in Europe (with not too high requirements). Thanks!
5/15/2017 11:16 PM
wronk :
francescacoo, it seems like neuroscience is one of several different topics, so it depends on your goals. Here are a couple more free and neuro focused resources that I'd definitely recommend (from personal experience).
https://www.coursera.org/learn/computational-neuroscience
http://courses.washington.edu/braindyn/ (This annual workshop will repeat in 2018)
5/16/2017 3:41 AM
dano :
Highly recommend that Computational Neuorscience course. I took it a few years ago. It provides a really interesting introduction and use case for differential equations
5/16/2017 6:36 AM
francescacoo :
wronk: thanks for the links! By the different topics you mean the others in computer science? It is possible to select one stream to follow and I would do the computational neuroscience one.. thank you! :)
5/16/2017 12:59 PM
braden :
Could anybody point me in the direction of the UX team Interaxon? I would love to get in contact with them to figure out what UX questions they�re trying to tackle. I�m a master�s student at Indiana University studying UX Design/UX Research, and I have a couple muse headsets that are looking for some action 🙂 thanks!
5/16/2017 1:51 PM
bryan_j :
Hahaha we should remember where...
5/16/2017 1:51 PM
bryan_j :
Jk
5/16/2017 2:08 PM
masha :
^^ cc graeme might be able to intro you to the right folks 🙂
5/16/2017 2:13 PM
yannick :
or hubertjb 🙂
5/16/2017 2:35 PM
braden :
I appreciate the leads! 🤓
5/19/2017 5:46 PM
dano :
Welcome to you all new people! 🙌
5/19/2017 6:48 PM
yannick :
Hey tonyb, pbonac, marcel.farres, brilhana & samuelbg13, please introduce yourselves in <#C08QEC5H7|introductions> so we can know a little bit more about you 🙂 And also look for your City channel in the list (e.g. <#C0VTC19DW|_toronto>) to meet other people from your local area.
5/19/2017 6:55 PM
bryan_j :
What what! Come join is in <#C0VTC19DW|_toronto>!
5/19/2017 7:03 PM
yannick :
Nan, join us in <#C0VFP9ST0|_montreal> 🙂
5/20/2017 2:01 AM
yannick :
<!channel>: For all the new comers (welcome!), you�ve probably received a message from our bot with some useful information. I�d like to re-iterate on subscribing to the Newsletter to stay updated on NeuroTechX activity (and the field of NeuroTechnology in general). As you might know we have multiple initiatives in motion - some of them just getting off the ground - so the next newsletter will be full of good stuff! You don�t want to miss it. :ntxblack:
I wish you all a great weekend 🙂
5/20/2017 3:20 PM
niouby :
hi everybody, my name is morgane and I am a french student. For a project with my school, I have got a MUSE and I tried the EEG101 application and it's very interesting!
5/20/2017 3:21 PM
niouby :
I have questions about the code of the application, am I on the right channel?
5/20/2017 3:25 PM
bryan_j :
dano could help you out with this!
5/20/2017 3:28 PM
niouby :
oh thanks, and where can I found him? (sorry but it's my first time on slack and I don't really understand how it works)
5/20/2017 3:29 PM
bryan_j :
No worries whatsoever! He should be able to respond directly to this thread! :) you may also send him a direct message by clicking his profile above and selecting the message/chat icon.
5/20/2017 3:38 PM
niouby :
thanks! I sent him a message but I think he is offline, I'll wait until he come back
5/20/2017 3:44 PM
bryan_j :
Great! He is quite active on Slack so I can imagine that he will respond within the day.
5/20/2017 5:29 PM
octonomy :
<#C08T2SENQ|devices> is better channel actually
5/20/2017 5:29 PM
octonomy :
Hi Morgane and welcome. We also have an <#C08QEC5H7|introductions> channel for saying hello
5/21/2017 7:26 AM
niouby :
Hi octonomy, thanks for the information 🙂
5/22/2017 9:26 PM
frawley_andrew :
Hi everyone,
I am looking to up my game on the technical side of neuroscience with the intent of working on early stage neurotech hardware. I have a business background but have been a deep enthusiast of neuroscience/neurotech for years. I understand concepts etc. but that is nil for building anything new.
I have three quick questions.
What foundational subjects should I be investigating first? (bio, chem etc.)
What resources do you recommend learning those subjects?
Is graduate school a logical step down the road?
Thank you to all who take the time to consider this.
5/22/2017 11:55 PM
neurotechreports :
Hi all, I'm new to this network but wanted to let everyone know about some progressive meetings on the neural engineering side. Cleveland Neural Engineering Workshop will take place June 23-25, 2017. It is not a conference but round table collaboration to build the future path for neural engineering with all the stakeholders. Here is a link to the Neural Engineering Roadmap, draft from the 2015 meeting: http://www.clevelandnew.org/images/resized/Informational-PDF's/Neural-Engineering-Roadmap.pdf
5/23/2017 12:05 PM
yannick :
Do we have anyone from Neuroon here? I decided to play the Neuro-Police this morning and I appreciate the reaction. Thumbs up.
5/23/2017 12:55 PM
dano :
frawley_andrew: I'd suggest making sure you�pick up some computational courses while in school: physics, calc, computer science, signal processing. That kind of stuff pops up really frequently in Neurotech.
5/23/2017 5:05 PM
frawley_andrew :
Hey Dano, thanks for the feedback. I am currently out of school and working in SF.
What do you think should be my first priority in learning?
5/24/2017 5:49 PM
bciguy :
yannick: Yes we are already offering this to our testers. Also, unless im mistaken, isnt this the same company that claimed you could get by on 2 hours of sleep &
5/25/2017 11:55 PM
yannick :
New Hearable (or earwearable as on their twitter) about to hit the market in few days:
�Wearable for Stress Relief & Brain Training�
https://comingsoon-tech.com/brainno-brainwave-heart-rate-wearable#
https://www.youtube.com/watch?v=7XgluoopDVI
Thoughts?
5/26/2017 12:18 AM
marion :
Nice! Hopefully they deliver on this. The Aware is another ear EEG device and unfortunately the kickstater project stopped after they raises their money http://spectrum.ieee.org/the-human-os/biomedical/devices/wireless-earbuds-will-record-your-eeg-send-brainwave-data-to-your-phone
5/26/2017 12:19 AM
marion :
I'd love to see the same form factor but with the battery not on the ear like the bose noise canceling earphones
5/26/2017 12:20 AM
marion :
5/26/2017 12:28 AM
yannick :
We have people from Aware in here.
5/26/2017 12:32 AM
yannick :
eohan ^
5/26/2017 12:38 AM
marion :
eohan is the kickstater still happening? People online seem to say it's over. The design of the Aware is great and I'd love to buy one.
5/26/2017 1:34 AM
marion :
Anyone else know if we can purchase the Aware somewhere?
5/26/2017 2:51 PM
rohit :
https://documents.epfl.ch/users/c/ch/chavarri/www/smc2017/smc2017_hackathon.html
5/27/2017 9:09 AM
harelasaf :
I'd also add that there are great MOOCs available. Such as Neuro Anatomy from Duke, computational neuroscience from Washington State University, Stanford has an especially popular machine learning class - all for free on Coursera.
5/27/2017 9:02 PM
francescacoo :
hey guys, just saw this, but not sure where to post it 😛 it's in Russia IEEE Brain Data Bank Challenge 2017
Brain Signal � Visualization & Analytics https://spcn.co/brain-initiative/?lang=en
5/31/2017 8:45 AM
stephen :
5/31/2017 8:46 AM
stephen :
Bryan�s talk at the Milken conference
6/1/2017 10:19 AM
francescacoo :
Hi Again not sure where to post this.. but it seems interesting (lab visits in Europe): "Eager to gain some practical experience?
Lab visits at HBP partner labs are a possibility for applicants to develop new technical or conceptual expertise for implementation in your research as well as your home institute's research program.
Further information and a list of contacts can be found on our website: https://education.humanbrainproject.eu/web/lab-visits"
6/1/2017 10:54 PM
wronk :
francescacoo, cool! Perhaps post it in the <#C0FNHPCF4|opportunities> channel?
6/2/2017 12:57 AM
sydneyneurotechx :
Yes please post in the <#C0FNHPCF4|opportunities> channel francescacoo 🙂
6/2/2017 10:17 AM
tonyb :
Hello all! I'm currently creating a VR meditation app, and looking to conduct research on users with the Muse headband. Anyone have experience with this? Looking for recommendations with research tools, protocols, etc. Cheers.
6/2/2017 11:04 AM
yannick :
You can ask more specific questions in <#C08T2SENQ|devices>, lots of people in here do have experience with the Muse headband 🙂
6/2/2017 11:41 AM
tonyb :
Thanks yannick
6/6/2017 9:10 AM
endoxaneuro :
Hello all, any feedback on the Halo Neuroscience? https://www.haloneuro.com what do you think about it?
6/6/2017 9:08 PM
francescacoo :
hi everybody, not sure where to post (as usual 😄 ). So, I am studying for a Msc in Software Engineering and it's time for my thesis. I would really love to do something related to BCI, as I have a great passion to learn more in the field. I have no academic background in neuroscience nor similar, I am not that strong at machine learning etc etc. We have a couple of thesis advisors which work on artificial intelligence and might be interested. The problem is that I have to submit an idea paper, and I am not sure where to start and what I could do!
So basically I was looking if you had any suggestions on where to take hints, ideas in which direction to move. I have a Muse headset and I was thinking to do something with that. But again, I don't have enough experience to understand if I could do something useful for a thesis!
6/6/2017 9:28 PM
stephen :
Who would you talk to and which unusual data sources would you include if you wanted to build the ultimate �quantified self� system?
As many types of data as possible, especially related to mental health. Wearables, time tracking, behavioral questionnaires, genetics, etc would all be possibilities?
6/6/2017 10:16 PM
dano :
francescacoo, I think one of the coolest things in the field rn is the possibility that machine learning could boost the efficacy of affordable consumer EEG to the point where it can provide such good benefits that people can't ignore it. IMU, what ML can do a really good job of in EEG is identify features and tailor itself to a user's particular brain. Because applying ML is relatively simple nowadays, I think figuring out whether that'll work at scale is really a software engineering question.
6/6/2017 10:18 PM
dano :
With Cloudbrain + Neurodoro, two NTX projects, we're looking into an architecture where neural nets can be trained online with huge datasets, but tweaked on a person's phone for their particular data
6/6/2017 10:21 PM
dano :
Google's announced that they're going to make Tensorflow able to run both on a phone and in the cloud and share weights https://www.androidheadlines.com/2017/05/tensorflow-lite-is-googles-optimized-tensorflow-for-android.html
6/6/2017 10:22 PM
dano :
So, there's an idea! Haha
6/6/2017 11:09 PM
stephen :
dano Where could I learn more about the Neurodoro project?
6/8/2017 4:09 AM
dano :
francescacoo https://research.googleblog.com/2017/04/federated-learning-collaborative.html?m=1
6/8/2017 7:26 AM
mhough :
hi endoxaneuro well they would definitely get a better "review" if they didn't make it sound like the military, Olympics, NBA, MLB, and NFL sports "medicine" people know what the hell they are doing and that this has been shown effective to people who should care. If I am wrong about that please post some links to peer-reviewed research articles etc. that I definitely missed. Even with those, and I take it they exist, check the reproducibility problem in neuroscience. Yes I understand the principle they are using. In a way, its kind of the opposite of sewing monkey's fingers together. Not exactly, but I am going to stick with that image. That the military can actually waste money on technology that doesn't work is really just common sense. Hope that helps
6/8/2017 7:37 AM
mhough :
I really do mean that honestly that it helps. These people should expect to get laughed out of a room. Not because what they are trying to do is bad but if you say you have a magic wand then show me some friggin magic and then teach me the incantation and make it public so it can get the review it needs. The fact that they don't should scare them - hopefully wise them up - nothing else
6/8/2017 7:48 AM
francescacoo :
Hey Dano,
6/8/2017 1:39 PM
francescacoo :
Hi everybody, a bit late, but there is a webinar starting soon "fNIRS and Brain-Computer Interfaces for Communication. An essential webinar for all BCI researchers: fNIRS enables the read out of voluntarily controlled brain states in real time and opens a communication channel with completely locked-in patients, otherwise isolated from the external world."
http://insidescientific.com/webinar/fNIRS-and-brain-computer-interface-for-communication-nirx
6/8/2017 3:53 PM
mhough :
Hi francescacoo yeah I should have said something too. It was pretty basic overview stuff although the application to ALS was nice to see. The recording will be online so you haven't missed it if your interested
6/8/2017 4:01 PM
francescacoo :
I watched until 15 minutes from the end.. it was good for me because basic :) so I could follow! yep the videos of what they have done was interesting!
6/8/2017 4:19 PM
davidevaleriani :
Yeah it wasn't that bad, the Plos Biology paper of the second speaker got quite a lot of attention when it got published actually
6/9/2017 1:41 PM
alirezad :
Hello guys!
6/9/2017 1:44 PM
yannick :
Hey alirezad, welcome on board, feel free to introduce yourself in <#C08QEC5H7|introductions>. Tell us more about you (where you are from, background, interests, how you found NeuroTechX&) :ntxt:
6/9/2017 2:02 PM
alirezad :
As a student who wants to continue studying PHD, I was wondering how is the job opportunities for someone with a PHD degree in BCI? (compared to someone with a machine learning PHD)
6/9/2017 2:05 PM
alexandre.barachant :
There is for sure more opportunity for someone with a PHD in machine learning. but my guess is that it is an ultra competitive field
6/9/2017 2:16 PM
alexandre.barachant :
or saying it differently, if you go on the Deep learning road, you better pick a topic that google and facebook and other are not going to solve within the next 5 years
6/9/2017 2:19 PM
alirezad :
alexandre.barachant Or maybe BCI research will be stoped by the united nation because of its dangers for human beings 😄
6/9/2017 2:22 PM
alexandre.barachant :
Unless you are working on invasive BCI in a lab with a very good reputation, BCI research funding is quite weak right now
6/9/2017 2:32 PM
davidevaleriani :
alirezad check the Connectionists mailing list, there are quite a few opportunities on neuroscience related topics for PhDs...
6/9/2017 5:51 PM
yannick :
Hey francescacoo, davidevaleriani & mhough, you guys were talking about fNIRS Webinar.
It is now accessible (same link): http://insidescientific.com/webinar/fNIRS-and-brain-computer-interface-for-communication-nirx
6/9/2017 5:52 PM
mhough :
Yep
6/9/2017 5:56 PM
davidevaleriani :
yannick: yeah, I've received the notification via email, but I've already watched it :) I think they could be useful for those who missed it, though
6/11/2017 7:59 PM
yannick :
We have been experimenting a little bit with the bot recently. If you have received multiple messages to the point that it feels like a bug / spam-ish, please let me know. It shouldn�t have happened, but if something slipped through, please let me know 🙂
6/12/2017 4:11 PM
samsagaz :
Hi guys
6/12/2017 4:20 PM
yannick :
Hey samsagaz. Where are you from? You can introduce yourself in <#C08QEC5H7|introductions> so we learn a little bit more about yourself
6/12/2017 9:12 PM
samsagaz :
yannick I'm from Argentina. Already introduced myself a few minutes ago.
6/12/2017 9:20 PM
yrenard :
samsagaz don't mind yannick, he is a bot and asks any new login in <#C073WP1T9|general> to post a short intro in <#C08QEC5H7|introductions> 😉
6/12/2017 9:47 PM
samsagaz :
Oh. Im feeling like a fool :)
6/13/2017 12:58 AM
watson :
Hey team, I just invited a friend who is a neurosurgeon doing brain implants (mostly DBS to combat parkinsons). I�m curious if there are any channels here that are most appropriate for his interests?
6/13/2017 1:03 AM
sydneyneurotechx :
watson: I'm assuming his interest is in DBS? Nothing yet. Should we create one? I wonder how many people would be interested.
6/13/2017 1:05 AM
watson :
I think his interest is in unpacking the brain�s mechanisms in general and understanding the potential from brain enhancement (e.g. Kernal)
6/13/2017 1:06 AM
watson :
(but I was asking about stuff that would be more directly related to invasive / implantable devices) Maybe just <#C08T2SENQ|devices> ?
6/13/2017 1:08 AM
sydneyneurotechx :
Posted a poll. Let's see what people think
6/13/2017 1:10 AM
sydneyneurotechx :
<!channel> Would you be interested in an Invasive/Implantable Devices Channel (ie what Kernel and neuralink are working on)? Please let us know in the poll above ^
6/13/2017 1:11 AM
c00p3r :
i would be interested, in learning more...
6/13/2017 1:23 AM
melanie :
sydneyneurotechx for clinical or non-clinical use?
6/13/2017 1:25 AM
yrenard :
melanie don't you believe it should be clinical-use only ?
6/13/2017 1:26 AM
yrenard :
what non clinical application do you oversee would use invasive / implantable devices ?
6/13/2017 1:31 AM
sydneyneurotechx :
Nothing DIY if that's what you're worried about melanie 😄
6/13/2017 1:32 AM
sydneyneurotechx :
I think there is just a lot of people interested in knowing more about the field
6/13/2017 2:07 AM
yannick :
During HackNight, we could experiment on how to open someone�s skull. I�ve seen many videos, I feel confident that I could pull this off. Trust me I�m an engineer!
Then we could compare the SNR of the OpenBCI when using electrodes on the scalp and on the brain.
Not much literature out there comparing the SNR of the same electrodes in the brain and outside the skull simultaneously. 🙂
6/13/2017 2:09 AM
benjamindeleener :
�Trust me I�m an engineer!� 🚀 😅 I am too, let�s do it
6/13/2017 2:18 AM
mrkrause :
There is, actually (though mostly in animals). It�s surprisingly similar, though you lose a lot of spatial specificity because of the skull and scalp!
6/13/2017 2:23 AM
sydneyneurotechx :
Alright the answer is looking pretty unanimous. <!channel> head on over to <#C5SJP8FC3|implants> for related convos 🙂
6/13/2017 9:49 AM
catpitch :
yannick: Bot worked great for me!
6/13/2017 12:00 PM
eferdinand :
If you didn't grabs yours yet, you have until the end of June to receive a devkit at http://neuroservo.com|neuroservo.com use promo code NTXJUNE2017
6/13/2017 12:38 PM
rohit :
eferdinand: not able to apply this Promo code
6/13/2017 12:58 PM
eferdinand :
ntremblay_neuroservo can you check this please
6/13/2017 1:00 PM
rohit :
I guess maybe some region based issue,no free stuff in India :P
6/13/2017 1:12 PM
yannick :
Please join me in wishing happy birthday to davidevaleriani who�s been a strong pillar of the NeuroTechX community through his involvement in NeuroTechLDN, NeuroTechEDU and also in exploring the neurotech opportunities and recruitment services NTX could offer. He also very recently received his �Dr.� title (finished his PhD) so I expect him to celebrate accordingly!
6/13/2017 1:53 PM
ntremblay_neuroservo :
Hello Rohit, you need to buy the device to receive the free dev kit so during checkout, you add a device, then a devkit, then you apply the promo code to have the discount applied. Let me know if you still have problem at checkout.
Nicolas.
6/13/2017 2:03 PM
davidevaleriani :
haha thank you NTX folks 🙂
6/13/2017 2:30 PM
sydneyneurotechx :
yannick: Happy Birthday davidevaleriani !
6/13/2017 4:29 PM
bryan_j :
Happy Birthday! Any big plans davidevaleriani??
6/13/2017 4:30 PM
bryan_j :
yannick: story of the doctor who implanted himself in South America comes to... mind. We have precedent!
6/13/2017 4:31 PM
davidevaleriani :
Thanks guys, a few things planned for the weekend, yes 🙂
6/13/2017 5:21 PM
buanzo :
davidevaleriani happy birthday!
6/13/2017 5:54 PM
samsagaz :
Didn't know about neuroservo! Looks nice.
6/13/2017 5:54 PM
samsagaz :
Btw, openbci devices are the best to start developing apps and start learning about bci?
6/13/2017 11:40 PM
lait :
the languages options are switched in the neuroservo webpage 😂
6/14/2017 4:08 PM
graeme :
Are we allowing marketing and selling now in the NTX general channel yannick?
6/14/2017 4:18 PM
graeme :
Not that this is a good idea to compare in a live subject, but you need different impedances inside and outside the skull.
6/14/2017 5:01 PM
aj :
Hey! Starting hacknights for New York City at the OpenBCI lab from 7-9pm. OpenBCI's on deck and there are a bunch of other awesome other electronics to hack on!
6/14/2017 8:39 PM
grandmasterspock :
Is there anyone on this channel who lives in the salt lake city area? (Utah, USA)
6/15/2017 4:17 PM
rgbn :
Hi everybody!
Does anybody have experience with ultrahaptics technology?
This looks like nice way do deliver stimuli for tactile BCIs and stuff like that.
6/15/2017 5:28 PM
yrenard :
rgbn I have tried Ultrahaptics in the past, it's outstanding piece of technology, but I would not recommend using it for tactile BCI based on my past experience on such BCI. Ultrahaptics gives you a sense of where in space the surface is located, but it certainly would not generate a brain response that compares to a vibrotactile stimulator because the stimulation would be much weaker. Did you have a chance to try some of their demo to get a taste of how that feels ?
6/15/2017 6:11 PM
alexandre.barachant :
rgbn in the past, i used LRA vibration motor to deliver vibrotactile stimulation : https://www.precisionmicrodrives.com/vibration-motors/precision-haptic-tm-haptic-feedback-vibration-motors
6/15/2017 6:16 PM
yrenard :
would definitely trigger a much stronger brain response I think
6/15/2017 7:17 PM
rgbn :
Yeah, I have just ordered LRAs from precision haptic, they just didn't ship yet.
6/15/2017 7:22 PM
rgbn :
yrenard: I don't know about any kits in Russia at all. Tactile BCIs generally don't work very good anyway, so if the stimuli is not strong enough there probably is no point in using it.
6/15/2017 9:30 PM
aj :
I just bought a couple of these!!
6/15/2017 9:33 PM
alexandre.barachant :
Adafruit has some nice breakout board for driving them. Otherwise you can use a PWM pin on an arduino (but pick the resonant frequency of the LRA)
6/15/2017 9:34 PM
alexandre.barachant :
i used them to do somato-sensory stimulation in children
6/16/2017 5:28 AM
eamon :
Gives a new meaning to "hacking"
6/19/2017 2:18 AM
dano :
Coverage of the Muse plugin for HTC Vive
https://medium.com/@micah31/what-eeg-can-bring-to-your-vr-experience-d2931652c4a8
6/19/2017 3:03 PM
aj :
sick@
6/23/2017 2:19 PM
buanzo :
awesome
6/24/2017 12:02 AM
a.tech :
Just a heads that we will be livestreaming a talk about quantum physics and neurobiology this Sunday.
Here is the livestream link for June 25th 3:30 pm PST. https://www.youtube.com/watch?v=NFC4g3t9Bys
More details at https://www.meetup.com/NeuroTechVAN/events/240659251/
6/24/2017 2:53 AM
dano :
Please record it and keep it up on Youtube if you guys can. I don't think I'll have time to catch the stream on Sunday, but I do want to watch
6/24/2017 1:07 PM
manonb :
Hi,
I need to add to the standard Muse electrodes a couple of additional electrodes.
Would someone be interested in doing this and selling me a pair? The technical documentation of the muse states:
"Auxiliary electrodes (enabled via [Settings] > [Show Aux Input]) can be connected on the Muse micro USB connector, by pin 4 on the MU-02 (BlueTooth LE) model."
http://www.musemonitor.com/Technical_Manual.php
6/25/2017 12:05 AM
rashi :
Is there a good Neuroscience Podcast?
6/25/2017 5:11 AM
pierre :
rashi Neuropod is good: http://www.nature.com/neurosci/neuropod/index.html
6/25/2017 5:12 AM
pierre :
this one is good too, but if you have time: http://snrp.utsa.edu/Podcast/Podcast.html
6/25/2017 5:12 AM
rashi :
pierre "Please note that from December 2015, NeuroPod is not being updated."
6/25/2017 5:13 AM
pierre :
ah too bad... I was thinking I haven't been listening to it in a while and I guess that's why hahah
6/25/2017 5:14 AM
pierre :
still recommend delving into the older episodes, as they were quite good and still pretty current
6/26/2017 2:41 PM
javi_neuralcubes :
We're trying give a push to the <#C1G552DD4|smartphone-bci> project, now known as icibici (pronounced as easy-bisi) for that we're going to order some pcbs of the project in bulk. We are wondering if anyone would be interested in getting some for organising hackathons for their chapters, or just to play around a bit. Also we're trying to get some funding to give the project a bit of push! join our channel if you want learn more!
6/26/2017 3:40 PM
sid_fulminds :
javi_neuralcubes: Hi Javi! could you please tell us a little bit more about your project?
6/26/2017 3:56 PM
javi_neuralcubes :
sid_fulminds yes this is the whole desc https://jmanart.github.io/smartphone-bci-hardware/
6/26/2017 3:58 PM
sid_fulminds :
nice! 🙂 thanks
6/26/2017 7:12 PM
dano :
I've never really come across a good neuroscience podcast after NeuroPod (RIP)
6/26/2017 7:13 PM
dano :
You might want to try the Super Science Happy Hour back catalog https://itunes.apple.com/us/podcast/the-super-science-happy-hour-with-matt-matt/id653573048?mt=2
6/26/2017 7:13 PM
dano :
They're two really geeky, kind of funny neuro postdocs who ramble. Some times they talk about their science
6/27/2017 12:17 AM
joeyo :
6/27/2017 12:18 AM
joeyo :
"Neuroscientists Talk Shop is the University of Texas at San Antonio's (UTSA) Neurobiology Podcast, showcasing the current research of internationally renowned guest Neuroscientists."
6/27/2017 12:18 AM
joeyo :
Ah Pierre already posted it, pardon
6/27/2017 12:23 AM
hfork :
Hey there, just joined the group.
6/27/2017 12:49 AM
gg :
hfork: Hey hey � welcome! 👋
6/27/2017 1:47 AM
dano :
Welcome, mehdi!
6/27/2017 1:47 AM
dano :
Tell us a bit about yourself
6/27/2017 2:40 AM
sydneyneurotechx :
hey hfork. Please introduce yourself in the <#C08QEC5H7|introductions> channel
6/28/2017 2:00 AM
bryan_j :
dano: AWESOMEEE - excited to see it in action this Friday!
6/28/2017 2:00 AM
bryan_j :
dano: Agreed!
6/28/2017 2:12 AM
fred7272 :
This presentation at the GPU Technology Conference is a bit computational in nature, but it seems like a significant improvement (from 4 hours to interactive) for this brain simulation tool. http://on-demand.gputechconf.com/gtc/2017/video/s7342-zigon-gpu-data-mining-neuroimaging-genomics.mp4 I am more of a general machine learning person, so I am curious to hear from computational neuroscience people. Is this best in class or there are better brain mapping/simulation tools out there?
6/28/2017 5:33 AM
sydneyneurotechx :
The Research was done in 2016, but nonetheless a cool article.
https://futurism.com/world-first-ultrasound-used-to-jump-start-patients-brain-out-of-a-coma/
6/28/2017 8:50 PM
griselda_pages :
Hi <!channel>, we're working hard on the re-design of the NeuroTechX's website. After having gathered some insights from the community, worked on strategy and designed the wireframes of the main pages, we're ready to start the design phase. But before that, we'd love to have your feedback:
>Is there anything missing? For example: user goals, pages, sections, functionalities...
>Is the document "understandable"? Would you change anything? Add anything?
Here's the doc that we're going to share with the designer: https://docs.google.com/presentation/d/1Q2zlV6bufMgW5bidtCcYKBVrRAsPnvoJe3h61Ym-9fQ/edit?usp=sharing
6/28/2017 10:47 PM
mhough :
sydneyneurotechx interesting article on ultrasound. I still haven't made it to the foundation but I will have a look. Don't know why they have such a nice tractography image but we will see:) Thanks for pointing this out.
6/29/2017 3:24 AM
sydneyneurotechx :
mhough: Yea, ultrasound still is one of those things that still needs a lot of exploration. As a Neuroimaging approach, ultrasound provide a better spatial resolution at a potential low cost.
Ref:
http://fultrasound.eu/clinical-neuroimaging/
https://maori.geek.nz/why-are-ultrasound-machines-so-expensive-623ce91d8402
6/29/2017 6:25 AM
mhough :
sydneyneurotechx thanks so much. Yeah familiar with all this. It's what Karl Pribrams partner would really like me to follow up on with the foundation in Charlottesville. It's a really big deal clinically. People interested in medical development should really be looking at this. Research wise my interest is in what you can do with sensor fusion or maybe better to say sensor synergy. Combining this with optical and EM is very rewarding but it might not be the best in terms of lowering the cost of a US system.
6/29/2017 4:49 PM
armin :
griselda_pages: Hi, can we have a section for scheduling local group meetings? For example, if it is possible that every three months those who are interested in Neurotech in Montreal have gathering to talk about Montreal news in Neurotech?
6/29/2017 4:53 PM
griselda_pages :
HI Armin! Thanks for the suggestion! But I'm not sure I understand what you mean. City chapters will keep organizing events via http://meetup.com|meetup.com, and those events are the ones that are going to be displayed in that "city" page. Also, all city chapters will keep their city channel on this slack, so the local community can chat in a more informal way, in between events. Does this answer your question?
6/29/2017 4:55 PM
armin :
Yes, thank you very much Griselda
6/29/2017 4:56 PM
griselda_pages :
you're welcome! Also, if anybody in the local community has any suggestions for future events, or activities, they can always contact the local organizers. They'll be more than happy to hear some ideas and get some feedback 🙂
6/30/2017 4:56 AM
joeyo :
6/30/2017 5:03 PM
tara :
Maybe link to calls to action earlier on the first section (seems to be the third screen down).
6/30/2017 6:54 PM
griselda_pages :
Hi tara! Thanks for your suggestion. Could you explain a little bit more please? What calls to action are u making reference to? And what section and page? Thanks!
7/1/2017 7:33 AM
kevin_ac :
Hi everybody,
Will someone go to EMBC 2017 that will take place on South Korea (Jeju) ? 🙂
7/1/2017 5:08 PM
tara :
griselda_pages I mean things like signing up for your newsletter/chapters and so on. You could have a join us button on the first screen/home page for instance and ask for email and location and match people up to chapters and events.
7/1/2017 5:09 PM
tara :
or just grab the IP location somehow.
7/2/2017 3:54 AM
ndartd :
anything on transcranial magnetic stimulation ?
I found this:
https://neurotechx.slack.com/archives/C1KFVAXCH/p1498421608471619
7/3/2017 10:54 PM
mhough :
Hi ndartd neuroelectrics is EEG/TES but I don't know who is making multicoil TMS. Not much has been done new with coils that I know of. Transmit/receive MRI coils are interesting. What are you looking for?
7/4/2017 1:00 PM
sheida :
Probably lots of eligible people here!
7/4/2017 3:04 PM
dano :
sheida: Hey Sheida! 🙂
Have any more info about Mobi? Looks like a new organization. Is it linked to a journal?
7/4/2017 4:13 PM
sheida :
Hey! First time I'm hearing about it too, looks like it's the first time there doing it! But I found out through Brain Products, which is what I use for my EEG studies so should be legit
7/4/2017 6:42 PM
dano :
Cool, so any papers published recently that have to do with "mobile brain imaging" are eligible for an award of <=$3000. Cool
7/7/2017 5:39 PM
fred-simard :
Thanks, good to know
7/7/2017 10:37 PM
stephen :
Who do you know that is interested in blockchain applications for handling health data?
cc andreacoravos watson pierre melanie
7/7/2017 10:55 PM
watson :
stephen: I�d recommend talking with my friend Alex Newman & not sure if he�s thought about that specifically but he knows about both: mailto:[email protected]|[email protected]
7/8/2017 7:04 PM
sydneyneurotechx :
There's a blockchain group here in Montreal that I know well, but they are mostly interested in the Fintech Vertical.
7/9/2017 5:23 AM
rashi :
Has anyone investigated hypnosis, subliminal messaging, consciousness and the like? I found a textbook on it and would love to discuss what I've read.
7/9/2017 5:24 AM
rashi :
Not to leave you hanging this is what I'm in the middle of
7/9/2017 5:25 AM
rashi :
7/9/2017 5:26 AM
rashi :
I put it in a text to speech app called @voice and went for a walk. It's a great way to exercise
7/9/2017 7:53 PM
lai_london :
rashi Hypnosis, and the associated areas seem interesting but Im always slightly skeptical of claims that are made, the link you provided looks interesting.
7/10/2017 3:22 AM
rashi :
lai_london I sort of cycle through believing and not, because there's not a whole lot of evidence either way. I've been able to consistently.
7/10/2017 3:22 AM
rashi :
Achieve name amnesia though
7/10/2017 3:25 AM
rashi :
7/10/2017 3:25 AM
rashi :
7/10/2017 3:25 AM
rashi :
7/10/2017 3:26 AM
rashi :
Very interesting and I enjoy their courses. I think they have a free one someplace
7/10/2017 3:33 AM
rashi :
Here's a free video that explains it well
7/10/2017 3:33 AM
rashi :
7/10/2017 3:39 AM
rashi :
I do something like this with the fingers most often
7/10/2017 3:39 AM
rashi :
7/10/2017 3:40 AM
rashi :
There's a bunch more on the channel about the rest
7/10/2017 2:52 PM
eferdinand :
7/10/2017 3:01 PM
joeyo :
https://www.engadget.com/2017/07/10/darpa-taps-five-organizations-to-develop-neural-interface-tech/
7/10/2017 8:54 PM
sydneyneurotechx :
related to joeyo 's post. The Darpa announcement http://www.darpa.mil/news-events/2017-07-10
7/11/2017 12:49 AM
yannick :
Anyone knows about Zeto? (http://zeto-inc.com/)
Wireless, Dry EEG headset. (that seems to target the medical/clinical market)
7/11/2017 5:33 PM
travjav :
Does anyone have any board for sale ( OpenBCI Gang.) Or something similar? thanks
7/11/2017 5:57 PM
samsagaz :
hey travjav im looking some to purchase too 🙂
7/11/2017 5:58 PM
samsagaz :
but dont know which one 😕
7/11/2017 7:29 PM
travjav :
samsagaz the gang. is nice! check out the official site for the specs and maybe ask others opinions how happy they are with their choice etc!
7/11/2017 7:30 PM
samsagaz :
ok 🙂
7/11/2017 7:57 PM
samsagaz :
guys, someoone know some android app that use BCI? i want to see what are the things that are made with BCI gears
7/12/2017 11:37 AM
jmhorschig :
yannick
I don't know them, but I found this:
"A compact, self-use EEG device that improves diagnosis and self-management of epilepsy"
I guess the 'medical-grade' is pure marketing for the US?
7/12/2017 2:18 PM
dano :
EEG 101 may be the only one right now https://play.google.com/store/apps/details?id=com.eeg_project&hl=en
7/15/2017 9:19 AM
francescacoo :
hey everybody, what about a "newbie" channel for people like me 😁 I have few questions on how to get started with a "project", some of them are very basic... I might post them here, but.. thanks!!! (you can't understand how much I would love to have chapter in Dublin to experiment with other people gggghhh)
7/15/2017 3:12 PM
yannick :
francescacoo you know that you are always welcome to ask your question in any channel. For example you can ask in <#C08T2SENQ|devices> for things related to the various devices or in <#C08PVFWH4|eegdata> if it�s a general question on EEG data.
We don�t �isolate newbies�. We welcome their questions in all and any channel as we were all newbies at one point. 😉
The recurrent questions are then captured and will be incorporated in NeuroTechEDU.
A good example of a recurrent question is �where can I find resources to learn on BCI stuff BCI?�, we then created the Awesome BCI List (https://github.com/NeuroTechX/awesome-bci) to answer that question.
7/15/2017 3:14 PM
yannick :
Another recurrent question is What are the current Neurotech Opportunities?, to which we are also preparing something exciting for the community.
(just dropping a teaser)
7/15/2017 6:56 PM
francescacoo :
Thanks yannick , I thought it could be good to see grouped the beginner questions in one place, but if you have already all these plans that's for sure good!
I will ask my questions then in the channels you have suggested, thank you!
7/16/2017 2:57 AM
dano :
Hey everybody,
EEG101 2.0 is in open beta. Download it here and try it with any Muse headset! https://play.google.com/apps/testing/com.eeg_project
7/16/2017 9:16 AM
francescacoo :
Neural oscillation: from an high-level perspective, so not looking for The Truth, is it correct to say that relaxation and light meditation correspond to an increase of alpha waves (and something of the tetha waves I have read). ? I know the topic is quite more advanced than this, I have researched a bit, but I'd like to have some more insight (I am using a Muse headset, so few sensors, and I am not looking to discover any new neuroscience thing, just analysing from the "software engineering" point of view for a thesis). Thanks!
7/16/2017 9:21 AM
alexandre.barachant :
jonas.chatel-goldman might have an answer for that
7/17/2017 6:12 AM
stephen :
Hey firmware people!
Do any of you have experience with the TI CC2541 BLE v1.4.2 stack and firmware image file formats (Standard, OAD, SBL) for the CC2541?
There�s a remote part-time or potential full-time opportunity for you with cool IoT /neurotech education startup in Santa Barbara!
7/17/2017 6:54 AM
jmhorschig :
francescacoo roughly speaking, yes. Alpha is higher though when being tired, so it's a bit questionable whether it's relaxation - the currently predominant hypothesis is that alpha represents inhibition of task irrelevant regions. That means, if you relax and do not require a certain brain region for that, you see an increase in alpha oscillations.
7/17/2017 9:20 AM
jonas.chatel-goldman :
@ francescacoo:
It really depends on how you define « relaxation » and « light mediation ». Tons of specific mediation practices exist that recruit different brain processes, it�s a bit like if we were discussing of a « light sport » activity without clarifying the type of exercice.
If you think of relaxation as simply having a low cognitive charge, then you may expect higher central-parietal alpha (e.g., Pz) and lower central-frontal delta-theta (e.g., Fz). This theta Fz/alpha Pz is a common ratio (e.g., Holm, A., Lukander, K., Korpela, J., Sallinen, M., & Müller, K. M. (2009). Estimating brain load from the EEG. The Scientific World Journal, 9, 639-651.)
Also, if you associate relaxation to a state with decreased vigilance (somewhat contradicts the « light meditation ») you may expect an increase in global alpha and in central frontal theta band (related to diurnal sleepiness and first stage of sleep).
Strijkstra, A. M., Beersma, D. G., Drayer, B., Halbesma, N., & Daan, S. (2003). Subjective sleepiness correlates negatively with global alpha (8�12 Hz) and positively with central frontal theta (4�8 Hz) frequencies in the human resting awake electroencephalogram. Neuroscience letters, 340(1), 17-20.
7/17/2017 9:31 AM
francescacoo :
Hi Jonas, thanks a lot.
I do understand that the reality is more complex than I thought.
From reading simple resources it looked like alpha were a good predictor for "relaxation".
My "goal" would be to try and determine (without any goal of making a new scientific discovery), which relaxation technique would be more helpful between few (let's say a breath guided meditation, relaxing music, body-scan meditation and so on - for people not expert in meditation I mean).
This is project for a software engineering thesis, so again, not trying to demonstrate anything or to create a commercial product, but more to explore and experiment classification algorithms and analysis of data.
I am aware that even commercial applications (as for example Muse), might use more complex analysis.
If I call it "relaxation" more than meditation do you think it could be more appropriate?
I will go and read some more on the topic, but any suggestion is absolutely welcome!
7/17/2017 7:37 PM
rouffi :
The first biannual Neuroadaptive Technology Conference, NAT'17, will be held in Berlin, 19th - 21st of July.
http://neuroadaptive.org/conference/2017
7/18/2017 12:41 AM
mhough :
rouffi you mean tomorrow:)
7/18/2017 12:46 AM
mhough :
It's heavy on former Makeig people but that ain't bad. Tim Mullen is great. I only know Makato from the mailing list
7/18/2017 10:42 AM
davidevaleriani :
Hey rouffi I'm going to NAT'17, are you? Anyone else from NTX?
7/18/2017 12:07 PM
rouffi :
Unfortunately no, I discovered this event last night #Serendipity
7/19/2017 6:03 AM
mhough :
davidevaleriani love to hear about it after if you have a chance to share
7/19/2017 10:36 AM
yannick :
Hey romy_lorenz, I saw you just started your postdoc, what will you be working on?
7/19/2017 5:49 PM
adamm_neurable :
Hey everyone! Is anyone attending SIGGRAPH by any chance? My company, Neurable, will be showing/demoing our BCI publicly for the first time. Would love for any of you to come by and try it!
7/19/2017 5:56 PM
alexandre.barachant :
adamm_neurable : That's great. Out of curiosity, what EEG device are you using ?
7/19/2017 6:03 PM
davidevaleriani :
7/19/2017 6:04 PM
yrenard :
SIGGRAPH is a great conference adamm_neurable , you'll have a lot of people flowing around, it is very crowded ! congrats for being part of the event 🙂
7/22/2017 1:00 PM
mhough :
adamm_neurable yrenard yeah siggraph is a hard core event where math is the product just as much as with cryptography. I'm planning to be there but it's right before ICON in Amsterdam.
7/22/2017 8:45 PM
adamm_neurable :
mhough do make sure to come through to Neurable's demo in VR Village!
7/22/2017 10:09 PM
aj :
adamm_neurable i wanna come! sneak me in!
7/23/2017 10:38 PM
bhargava2566302 :
working on new neurochatbot i will be updating the repo soon
7/25/2017 6:20 PM
lai_london :
is there any relative inexpensive technology that can be used to give a meaningful measure of executive function?
7/25/2017 6:20 PM
lai_london :
just asking so I can learn more
7/25/2017 6:22 PM
benbrain :
lai_london what do you mean by 'measure of executive function'?
7/25/2017 6:28 PM
lai_london :
I feel like I will probably end up conflating several more basic processes, but when I mean in everyday terms, is like attention or cognitive efficiency?
7/25/2017 6:29 PM
lai_london :
I know thats very vague, I was wondering if someone here could shed more light on this for me. e.g like depending on your mood at different times of the day you are more efficient at cognitive tasks,
7/25/2017 6:29 PM
lai_london :
I just wanted to know if there was anything that could measure that in some meaningful way.
7/25/2017 6:30 PM
lai_london :
I have heard about muse, but im still educating myself about the underlying technologies
7/25/2017 6:37 PM
benbrain :
lai_london EEG technology (like the Muse) is probably your best bet. However, you'll be measuring correlates of things like attention and cognitive load and not directly measuring those processes (presently we really have no clear idea of what "measuring" cognitive capacities would entail). There is a whole bunch of research that aims to identify EEG correlates of these types of things, so once you understand how EEG works and what it is measuring, you should start to dig into the nuances of that space.
7/25/2017 6:37 PM
lai_london :
thanks
7/25/2017 6:38 PM
lai_london :
really appreciate your answer
7/25/2017 6:38 PM
dano :
In the <#C4LSDAE95|neurodoro> project, we're working on throwing together an open source classifier for attention and cognitive performance with Muse and TensorFlow. You might be interested in that.
7/25/2017 6:39 PM
lai_london :
thanks
7/25/2017 6:39 PM
lai_london :
sounds great, would be good to meet someone that has a muse headset
7/25/2017 6:39 PM
lai_london :
I can think up some pretty interesting �studies� I would like to try out
7/25/2017 7:09 PM
mhough :
That's awesome dano
7/25/2017 10:13 PM
hectordomorozco :
lai_london In terms of the kind of analysis you can do, people have been using �Transfer Entropy� as a �cognitive measure�. Problem is, you wouldn�t be able to do that with a Muse. You�d need something like OpenBCI and at least (I�d say) 8 electrodes
7/26/2017 5:30 AM
yannick :
In case you missed it (because apparently I did)... Last month, Philips bought EGI (EEG).
7/26/2017 7:18 AM
yannick :
After chriscrawford recently becoming Dr. Crawford, join me in congratulating marvinandujar in successfully defending his PhD. The world of BCI just welcomed 2 new great Dr.
If you don�t know them, but have heard about the Brain Drone Race, then you know them 😛
7/26/2017 11:42 AM
okbalefthanded :
Congrats for both of you chriscrawford and marvinandujar , if you don't mind sharing your dissertations for curious readers like myself
7/26/2017 3:22 PM
chriscrawford :
Thanks!!!! yannick
7/26/2017 3:23 PM
chriscrawford :
okbalefthanded sure. Will share link.
7/27/2017 12:46 AM
marvinandujar :
yannick - Thanks!
7/27/2017 9:30 AM
nuno :
Congrats Dr. chriscrawford!! 😄
Would love to get the link as well! Thanks 🙂
7/27/2017 4:47 PM
tahiltonjr :
interested in communicating with anyone using hypnosis with eeg data or anyone interested in learning my research from the two
7/27/2017 11:36 PM
bryan_j :
Hi there NeuroTechX community,
I hope that you are well.
I have a favour to ask � my friend Tyler is a PhD candidate at the University of Toronto who is working on batteries for wearables, sensors, and other hardware devices. He is in the process of building a company, and would like 15-20 minutes of your time to determine the exact needs of the end user and any potential real-world problems that are � and may be � experienced.
This interaction could be a phone interview or a coffee in-person; whatever works best for you. If you are interested and would like to learn more, feel free to email Tyler at mailto:[email protected]|[email protected] to schedule a time to meet.
Thanks!
7/28/2017 7:24 PM
fred-simard :
yannick I hate to be the bearer of bad news, but I can't upload any more documents. It tells me the slack storage is full... I'm posting it here, because I think part of the solution will need to come from the community.
7/28/2017 8:58 PM
dano :
Yep. getting that message too
7/28/2017 8:58 PM
dano :
I've done my part to delete some of my old files, but unfortunately I've found it really difficult through the Slack interface. Is there any way to delete multiple files at once?
7/29/2017 4:06 AM
yannick :
fred-simard thanks for letting me know! I�m currently deleting my older files.
(stuff before 2017 that are not relevant anymore)
But you are right in saying that the solution will come from the community. If everyone could quickly go check their files and delete the older ones that are not relevant anymore and/or upload the big videos on G-Drive and share the link, as opposed to uploading the video in here.
I know that for some of you (aj, dano, alexandre.barachant, sydneyneurotechx, etc.) it might be more painful since you�ve transferred many files.
Indeed the Slack Interface for deleting files is sh**. Most likely on purpose to drive sales, because it would be very easy for them to make it as nice and easy to use as everything else they do.
If we all take 5-10 minutes, we�ll fix the problem in no time.
Thanks 🙂
Here is how to check your files: https://neurotechx.slack.com/files/ (and go to �My Files�)
7/29/2017 2:57 PM
sc :
Any progress on getting a non-profit license for Slack?
7/29/2017 2:57 PM
sc :
(that�s probably also why were weren�t in that directory listing posted awhile back)
7/29/2017 3:22 PM
fred-simard :
yannick Thanks for the details. I'll review my way of publishing my documents. I'll host them somewhere else and post the links instead. It should actually turn out a better solution overall anyway.
7/30/2017 6:00 PM
yrenard :
LOL
7/30/2017 7:58 PM
kiret :
i mean... it kinda IS a pain-inducing torture device if you're the researcher
7/31/2017 10:25 AM
yannick :
LOL& Ep
ioc.
7/31/2017 11:43 PM
yannick :
Hey bciguy, you�ve probably seen the article from elizastrickland in IEEE Spectrum 2 weeks ago about Lucid Dreaming Technologies.
With Dreem raising $22M (just for sleep, not for Lucid Dreaming), I�m quite curious about the evolution of the Sleep Wearable Devices.
8/2/2017 7:04 PM
bciguy :
Hey yannick and yes, it DOES feel good to be copied 🙂 in my opinion EEG is going to make the biggest consumer impact most quickly in the sleep space first. This is partially because the sleep EEG signals exhibit such high SNR but also because the modern world needs better solutions for energy management. For all intents and purposes, lucid dreaming is still an industry in its infancy (see my comments above) with many open questions yet to be addressed. Im pleased today to share with the nTX community (before anyone else!) that our platform was designed from the ground-up to provide answers to such questions as posed in the above article. You can read our new update here: https://www.kickstarter.com/projects/iwinks/the-aurora-dream-enhancing-headband/posts/1938439
8/2/2017 7:04 PM
bciguy :
Yes she brings up good points about the LD space in the title, but then goes on to assume lucid dreams are a commodity (which they are not.. yet) and that the Aurora offering is in �last place�, (when in reality Aurora is the only product mentioned, the others are campaigns ) Hers would be a fair expectation if the other groups highlighted in her article didnt have a track record of abandoning campaign promises. The truth is that we�re the only team who has remained 100% committed to the challenge of a safe lucid dream induction method as well as seeking real scientific evidence to prove its efficacy. For these reasons this was a difficult read, especially considering all that the Aurora team has fought for over the last 4 years.
8/3/2017 1:58 AM
andreacoravos :
8/3/2017 12:10 PM
eferdinand :
For who didn't read about upcoming hackathons http://www.br41n.io/
8/4/2017 7:27 PM
danielao :
Hi all, I don't know which channel I should ask this so I figured I'd ask here.
I haven't been that active on this slack but I'm currently looking at getting into basing work around mobile EEG. I want to integrate mobile EEG with self-reported data and eye-tracking to develop a profile on the engagement of a user on product (something that I know people are doing already in different capacities). I was wondering if anyone had any recommendations on headsets they enjoy using that work well with eye-tracking and are fairly intuitive with how the data and software work to produce usable results.
For background, I have used research EEG (the Brain Vision Suite) for an MSc thesis so I'm familiar with that process. Sorry for the long winded message. If this needs more clarifying, let me know. Thanks!
8/6/2017 3:26 PM
eferdinand :
danielao the abm b-alert live x10 is a good device easily integrates with the tobii.
I've been using both for quite a while now.
With some additional coding if not in real time you can do offline synchronization and analysis.
Let me know if you need more information via direct message or by email
8/8/2017 10:48 AM
kamitor :
Hello Everyone! 🙂 Happy to be here. My name is Kamitor! or Christiaan Verhoef. I'm based in the Netherlands and I'm here to help make this project boom ! 👍😁 Super excited
8/8/2017 12:35 PM
jmhorschig :
Welkom! What project? Neurotechx?
8/8/2017 1:47 PM
elizastrickland :
Hi all, thought you might be interested in my Q&A with Neurable's CEO about the EEG-controlled VR game the company unveiled at SIGGRAPH last week. Thanks to adamm_neurable for making the connection. http://spectrum.ieee.org/the-human-os/biomedical/bionics/brainy-startup-neurable-unveils-the-worlds-first-braincontrolled-vr-game
8/8/2017 3:38 PM
adamm_neurable :
And thanks to elizastrickland for the wonderful piece!
8/8/2017 4:12 PM
dano :
Amazing read!
8/9/2017 9:46 AM
kamitor :
Wow! Thats exactly what this needs too 😄
8/9/2017 4:45 PM
timopheym :
hey all, we are here in Moscow also playing with VR and EEG, and now we wanna to mix them, for BCI we are wanna use FreeEEG32 (https://hackaday.io/project/20618-freeeeg32-32-channels-electroencephalography) and for VR we have HTC vive, is anyone can suggest some suitable cap? Corse the openbci cortex that we was used before cannot be combined with vive =(
8/10/2017 4:00 AM
sc :
timopheym You can get a replacement strap for you Vive from about $20. Why not hack one of those to use electrodes with OpenBCI? Use a tape measure for accurate 10-20 placement on your head.
8/10/2017 7:23 AM
timopheym :
sc For our clients we need something that can be placed fast enough, in term of minutes.
8/11/2017 5:10 PM
spit :
Hi guys. I read about the icibici (https://icibici.github.io/site/). Is it possible to order the kit somewhere?
8/12/2017 10:34 AM
davidevaleriani :
Hey @split, check <#C1G552DD4|smartphone-bci> for that.
8/12/2017 10:54 AM
spit :
Thanks 🙂
8/15/2017 9:00 PM
kim :
Hi everyone!
Interaxon is launching the new SMITH Lowdown Focus soon, and we�re inviting all of you to our office to try it out! See how a pair of sunglasses can help you build focus 😄
https://www.meetup.com/Muse-Meditation-Meetup/events/242570232/
8/15/2017 9:36 PM
kim :
oh probably should mention that these sun glasses detect EEG
8/15/2017 10:57 PM
aj :
So stoked for these
8/15/2017 11:01 PM
a.tech :
kim not only will we look cool, but it will make sure that we feel cool 😎
8/15/2017 11:01 PM
a.tech :
can't wait to try!
8/17/2017 1:58 AM
sydneyneurotechx :
Set up a test location in Montreal 🙂
8/17/2017 5:27 AM
bechir :
http://www.nature.com/news/china-launches-brain-imaging-factory-1.22456
8/17/2017 3:30 PM
jamoonie94 :
I'm going to this event! Apparently my lab has a collaboration with Interaxon. Supposedly P300s can be detected with muse though I haven't looked into the tech yet
8/17/2017 7:24 PM
travjav :
Does anyone have any information on how to map the Microvolts for the Ganglion? I need to display the information accurately on a line graph with live data from a socket(s) thanks in advance !
8/19/2017 3:57 PM
graeme :
This is a good story on research into EEG in sports performance featuring portable headsets: http://www.cbc.ca/amp/1.4229868
8/19/2017 4:21 PM
dano :
Cool stuff
8/19/2017 4:22 PM
martys :
Woah. Batters up 👍
8/21/2017 11:54 AM
timopheym :
Hey guys! Is there someone who has develop openbci device? We did broken ours, and tried to change controller so there is some troubles, with changing firmware. Thanks and love!
8/21/2017 12:54 PM
aj :
timopheym me!
8/21/2017 12:54 PM
aj :
Please direct message me
8/21/2017 12:55 PM
aj :
If it's a bug we can open an issue or make a forum post
8/22/2017 11:44 PM
teon :
hey everyone, I was wondering if there is someone who can help with a
webpackproblem
8/23/2017 12:38 AM
aj :
What's the problem?
8/23/2017 12:39 AM
aj :
Is it from the openexp?
8/23/2017 3:45 AM
sydneyneurotechx :
yup! It's possible jamoonie94 . Check out alexandre.barachant 's blog post on it.
8/23/2017 1:51 PM
jamoonie94 :
Thanks! This is helpful. What are your thoughts on it's signal acquisition and quality? I read in one of the papers that ~5min use was the maximum due to varying head shapes. Something I am also concerned with is the delay and jitter due to Bluetooth.
8/23/2017 2:15 PM
alexandre.barachant :
jamoonie94 i'm not sure i understand the "maximum use due to varying head shapes" ? do you mean it hurt after 5 minutes ?
8/23/2017 3:52 PM
jamoonie94 :
In the sense that electrode contact is inconsistent within and between individuals
8/23/2017 4:15 PM
alexandre.barachant :
i would say it is as consistent and stable as any other EEG
8/23/2017 4:17 PM
alexandre.barachant :
your main concern with the muse should be more about the channel location with respect to the activity you want to probe.
8/23/2017 5:31 PM
dano :
Relevant to my interests. Does this have to do with the electron build of openexp? I think there's a fundamental issue using a web-designed bundler for a desktop app
8/23/2017 5:37 PM
dano :
Maybe Electron Forge instead? https://electronforge.io/
8/25/2017 3:03 PM
fred-simard :
I believe these people are part of the network: http://getcloudbrain.com/, I would like to get into a quick chat with them, to hear the what and how of their platform and discuss further matter.
8/25/2017 3:11 PM
fred-simard :
thanks!
8/28/2017 4:05 PM
marion :
Hey Fred! Pm me and we can chat more.
8/28/2017 5:13 PM
adamm_neurable :
Hey! We (Neurable) had a pretty cool feature today - https://www.nytimes.com/2017/08/27/technology/thought-control-virtual-reality.html?smid=tw-share Check us out!
8/28/2017 5:57 PM
maya :
super cool adamm_neurable!
8/29/2017 3:53 AM
sydneyneurotechx :
BCI and Virtual Avatar being used to help treat Gait disabilities - http://newspublicist.com/virtual-avatar-help-people-with-gait-disabilities/
8/29/2017 2:53 PM
danielao :
Even though I've done some investigating myself, I was wondering if anyone is aware of a source that says what different commercial EEGs are actually capable of reading or analysing (not just what marketing or news articles)?
8/29/2017 2:58 PM
yannick :
Hey danielao, jamoonie94 recently asked a similar question in <#C08T2SENQ|devices>. You might want to have a look.
8/29/2017 3:23 PM
alexandre.barachant :
danielao this is a tricky question, because most commercial EEG don't really offer analysis. Theoritically speaking they are very capable, but in practice there is not a lot of usefull metric that are exploitable out of the box
8/29/2017 4:20 PM
danielao :
Hey yannick. Thanks for the heads up. But alexandre.barachant touched upon what I was getting at. The descriptions under Muse and Emotiv say any paradigm that needs that amount of electrodes or less but I am looking for the translatable insights such as engagement, concentration, etc that can reliably be seen.
8/29/2017 4:23 PM
danielao :
Or I suppose at a basic level it comes down to where in the cortex is the point of interest and are there at least on or two electrodes that are mapped on or near that area but it would be nice to not have to figure that out in reverse from papers.
8/29/2017 5:45 PM
omer :
How it is even possibile to get reliable EEG during sport e.g running. I'm skeptical, have anyone recorded anything like that that was usable?
8/29/2017 5:48 PM
yrenard :
omer these are pretty old but might be interesting :
8/29/2017 5:49 PM
yrenard :
8/29/2017 5:50 PM
yrenard :
Also Wearable Sensing devices deliver outstanding measurements when moving around
8/29/2017 5:53 PM
omer :
I doubt that. I see why hardware companies will claim that and try to market such specific products. Roller coaster is not sport, if you sit still this might be possible, but I'm skeptical for the ping pong, as for spinning, I guess you could be trained not to move while cycling, but what about the sweat. What makes you say that last sentence?
8/29/2017 5:54 PM
omer :
I'm wondering if anyone could quote a reliable paper doing any EEG measurement while heavy sweating or while moving a lot...
8/29/2017 5:54 PM
yrenard :
that is the best I can quote without searching 😉
8/29/2017 5:54 PM
yrenard :
it is outstanding
8/29/2017 5:54 PM
yrenard :
for the last sentence, I have seen the data
8/29/2017 5:55 PM
omer :
That's cool, but these our adds...
8/29/2017 5:55 PM
omer :
*are ads
8/29/2017 5:55 PM
yrenard :
fair
8/29/2017 5:55 PM
omer :
I take your word, what kind of data?
8/29/2017 5:55 PM
yrenard :
raw EEG
8/29/2017 5:56 PM
omer :
while runnnig?
8/29/2017 5:56 PM
yrenard :
moving around was more walking
8/29/2017 5:57 PM
yrenard :
I haven't seen Wearable Device being used for running, and because it is dry electrodes, I would be worried that the hit to the ground creates motion artifacts that can not be recovered
8/29/2017 5:57 PM
yrenard :
but for walking, it is very good
8/29/2017 5:58 PM
omer :
walking like a patient or walking in the park?
8/29/2017 5:58 PM
yrenard :
what should I understand as the difference ?
8/29/2017 6:01 PM
omer :
like they way you walk for lunch comparing the way you 89 grandma walk
8/29/2017 6:02 PM
omer :
also, in the brainproduct add of the bicycle it actually says:
8/29/2017 6:02 PM
yrenard :
it was more about how I walk to go lunch
8/29/2017 6:02 PM
yrenard :
though sometimes I may run 😉
8/29/2017 6:02 PM
omer :
🙂
8/29/2017 6:04 PM
omer :
Anyway, if anyone has manage to recorded and analyze while heavy sweating and/or running I'm very curious to what extent the data is usable..
8/29/2017 6:37 PM
fred-simard :
With a bit of luck, walking/running artifacts might be removable with ICA. You might benefit from adding an accelerometer to the setup and feed it to the ICA as well, but that's hindsight.
8/29/2017 6:45 PM
yannick :
That�s the kind of things that MoBILab is trying to facilitate (https://sccn.ucsd.edu/wiki/MoBILAB)
Here are 2 cool videos of real-time artifacts removal.
https://www.youtube.com/watch?v=qYC_3SUxE-M
https://www.youtube.com/watch?v=wESFVDKHRBE
8/30/2017 2:06 PM
omer :
Thanks yannick ! that enlightening... Do you know if the algorithm in the videos is out there or they keep it private?
8/30/2017 2:47 PM
endoxaneuro :
Hello all, could you please read and if you like it 😉 please share my recent article https://www.weforum.org/agenda/2017/08/ai-and-human-stupidity
8/31/2017 1:09 AM
dano :
That artefact removal is insane
8/31/2017 1:39 AM
yannick :
Do we have people going and/or interested in going to BrainFutures?
9/1/2017 6:19 AM
zachb :
Any developers with Muse experience interested in some extra work? Please PM me if you are interested. Thanks!
9/1/2017 9:21 PM
mattherich :
Hey all, benjamindeleener, yannick, myself and others have been working diligently to get the NeuroTechX Student Clubs Initiative up and running and we're proud to announce that the first Q&A with PolyCortex has been published! https://medium.com/neurotechx/brain-powered-boat-race-q-a-with-polycortex-a5bd3e345f99
9/1/2017 9:22 PM
mattherich :
Stay tuned, we'll be publishing new Q&As with student clubs every other week on Fridays along with other NeuroTechX Initiatives in the future!
9/2/2017 10:57 AM
mikhail.sintsov :
I don't know if you've seen it or not, but still http://ncase.me/neurons/
9/2/2017 10:12 PM
graeme :
Hey Yannick I'm on a panel at BrainFutures. Should be a good (if small) conference. Last year's speakers were very good and this year's look even better.
9/4/2017 1:13 AM
yannick :
Hey omer regarding our recent discussion on EEG while walking/running& this might be interesting: http://www.sciencedirect.com/science/article/pii/S1053811917305815
( haven�t read it yet, but clearly added it to my ReadMe list 🙂 ) /CC yrenard
9/5/2017 5:02 AM
joshuad :
http://journals.plos.org/ploscompbiol/article/file?id=10.1371/journal.pcbi.1005268&type=printable
9/5/2017 8:49 AM
mikhail.sintsov :
I would suggest reading this at first, https://bml.bioe.uic.edu/BML/Stuff/Stuff_files/biologist%20fix%20radio.pdf
9/5/2017 9:35 AM
mikhail.sintsov :
There is already a reference though
9/5/2017 4:25 PM
yrenard :
brilliant, I love it
9/5/2017 4:51 PM
joshuad :
Ah yes, the original, thanks for the link mikhail.sintsov
9/6/2017 1:56 AM
yannick :
Congratulations to benjamindeleener who defended his PhD today! 🍾🎉🤘
Benjamin has been an active member of NTX community, specially in Montreal and he�s been leading the Student Club Initiative for 2017.
(and cancelling this week�s meeting, because he had �a PhD defense�& #FindOtherExcusesPlease)
9/6/2017 1:59 AM
benjamindeleener :
Thank you yannick! I promise you I will find other nice excuses to skip our weekly meetings :-D
9/6/2017 4:01 AM
davidevaleriani :
Congrats!!! 🎉
9/6/2017 3:11 PM
hectordomorozco :
Congrats dudie!
9/6/2017 3:52 PM
elizastrickland :
Hi everyone -- here's an article written by the team that won the BCI race in last year's Cybathlon. It talks about how they chose the brain signals for their BCI and how the racers trained with it. https://spectrum.ieee.org/biomedical/bionics/how-we-won-gold-in-the-cyborg-olympics-brain-race
9/14/2017 6:36 PM
psoulos :
Does anyone know if developers will be able to interface with the Muse X Smith sunglasses through the Muse SDK? http://www.smithoptics.com/us/lowdownfocus
9/14/2017 6:43 PM
yannick :
hubertjb ^ ?
9/16/2017 5:49 AM
dano :
You can connect to the Focus glasses with the current Muse SDK. At our last hacknight at the raxon office I paired EEG 101 to a pair of glasses across the room.
9/16/2017 9:47 AM
alexandre.barachant :
muse-lsl also work for this
9/20/2017 12:30 AM
sc :
Kickstarter for a new NeuroSky-based headset: https://www.kickstarter.com/projects/1571399980/neuroplus-improve-focus-with-brain-sensing-headset
9/20/2017 1:08 AM
aj :
Wow!
9/21/2017 7:33 PM
dano :
Hmmmmmmm :thinking_face:
9/22/2017 9:08 PM
sydneyneurotechx :
any good references/publications that benchmark consumer eeg's to research/medical grade?
9/22/2017 9:08 PM
sydneyneurotechx :
I found this one, but looking for others
9/25/2017 2:12 PM
danielao :
Also interested in this.
9/26/2017 2:23 PM
dmorden :
Nothing; most researchers say the data off of consumer is shit
9/28/2017 2:05 PM
keaton :
May have been posted before, but this is a really good paper on the history, tech, and current state of BMIs today. Long, but very thorough. http://physrev.physiology.org/content/97/2/767.abstract
9/28/2017 2:22 PM
ruudkalis :
Awesome thanks for sharing!!
9/28/2017 2:30 PM
keaton :
Oh, and just DM me if you can't access because of the paywall
9/28/2017 4:13 PM
yrenard :
hey keaton can you share the PDF to us ?
9/28/2017 6:36 PM
keaton :
Thanks, hailey!
9/29/2017 12:47 AM
jfrayshe :
Hey everyone, we�re looking to bring on contributors to continue the great work that has been happening with the Neurotech Edu project! If you want to be part of a platform that will educate budding neurotech enthusiasts we are looking for you! Content creators and field experts are welcome, message me jfrayshe or sydneyneurotechx to get involved!
9/29/2017 10:27 AM
alexdni :
Thanks hailey for this amazing treatise on BMI / BCI, some really comprehensive and fascinating stuff
9/30/2017 8:28 AM
ozzie :
paper <<Converting Your Thoughts to Texts: Enabling Brain Typing via Deep Feature Learning of EEG Signals>> https://arxiv.org/pdf/1709.08820.pdf
9/30/2017 9:35 AM
alexandre.barachant :
hum, interesting paper.
At first, performance seems amazing. then you realize the evaluation is deeply flawed.
They classify each EEG time sample, and they do a random split. Since consecutive EEG time sample are really correlated, and that they did not split based on time, the good performances can be basically explained by the RNN memorizing similar samples
9/30/2017 9:37 AM
alexandre.barachant :
this is why they do not share online results, and instead write "The accuracy in the online mode is however lower than what can be achieved in an offline setting"
9/30/2017 8:41 PM
yannick :
I understand the 1x64 input vector for the 64 channel device, classifying each EEG time sample.
But in their second exp. with the Epoc, they say the following:
�In our practical deployment, the sampling rate of Emotiv EPOC+ headset is set as 128Hz, which means the server can receive 128 EEG recordings each second. Since the brainwave signal varies rapidly and is very easy to be affected by noises, the EEG data stream is sent to server each half second, which means that the server receives 64 EEG samples each time. The 64 EEG samples are classified by the deep learning framework and generates 64 categories of intents.�
So they take 1 channel, 64 points? (half a sec)
Not sure if/how they handle multiple channels as input.
9/30/2017 8:42 PM
yannick :
Yeah, my red flag alert went off as well.
�lower�& How lower? Give me numbers!!!
10% is lower than 95%, and I�m not gonna read further 😛
9/30/2017 8:43 PM
yannick :
90% you�ll still have my attention. But please don�t hide these numbers& 😕
9/30/2017 8:51 PM
alexandre.barachant :
i think they just process them by batch i.e. receive 64 sample and produce 64 prediction (or categories of intents ?)
10/4/2017 7:27 PM
graeme :
Hey guys, the Smith Lowdown Focus glasses are now available for purchase. They work really well with MuseLSL. We'll be sharing some code and data from our internal tests on alexandre.barachant's MuseLSL repo over the coming weeks.
10/4/2017 7:27 PM
graeme :
10/6/2017 1:58 AM
llean :
Lucid Dreaming EEG + tDCS headset, In Demand: https://www.indiegogo.com/projects/lucidcatcher-dreaming-becomes-real-dream-sleep/
10/6/2017 2:09 PM
sanyaade :
Hello Everybody, Sanyaade from London
10/6/2017 6:27 PM
francescacoo :
Hi everyone! I am starting to work on my thesis (in software engineering). I am doing something which is related to relaxation/meditation and EEG (using Muse headset). I am still working to my literature review and reading different papers.. as there are many students here I was wondering if anyone knows of some great paper I should know / or has any suggestion for me I would really appreciate it! Thanks a lot!
10/6/2017 10:41 PM
dano :
Here's a nice pape for you, frances: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3837242/
10/6/2017 11:17 PM
francescacoo :
wohoo thanks a lot dano !!!
10/9/2017 4:29 AM
sheida :
Richie Davidson's work would be relevant: https://centerhealthyminds.org/science/publications
10/9/2017 9:20 AM
francescacoo :
fantastic I will check it now! Thanks a lot!
10/9/2017 5:51 PM
yannick :
How was the conference and the Hackathon? (CC falk, aj and others)
10/9/2017 5:52 PM
yannick :
(SMC 2017 in Banff)
10/9/2017 9:36 PM
aj :
A blast!! Also exhausting
10/9/2017 9:36 PM
aj :
A bunch of teams used openbci and majority had success and won cash
10/9/2017 9:37 PM
aj :
I learned a lot of lessons and was able to get a nice stability patch for wifi shield firmware for the raw top JSON conversion and through mqtt
10/9/2017 9:38 PM
falk :
Blast indeed! Check out the MuSAE Lab twitter feed to see the photos of the winners. Video of the team presentations should go up on our youtube channel soon& stay tuned!
10/9/2017 9:39 PM
aj :
Great job Tiago falk !!
10/9/2017 9:40 PM
falk :
thanks, but Christoph and his G.tech team did all the work! And thanks to the sponsors who made it happen& without the equipment, there�d be no hackathon!
10/9/2017 10:10 PM
yannick :
Awesome!
10/12/2017 1:14 PM
urish :
Just published - my adventures connecting Muse 2016 to the web!
https://medium.com/@urish/reactive-brain-waves-af07864bb7d4
10/12/2017 8:29 PM
elsehow :
hey folks. just got the lowdown focus from smith optical/interaxon. anyone have any luck pairing it with
muse-io?
10/13/2017 10:34 PM
elsehow :
update: heard from graeme - I got a BLE adapter from amazon. will update folks as I learn more.
10/17/2017 3:57 AM
alexander.chudinov :
Newbie Question Incoming!! Do I only need an electrode hooked up to a micro controller to measure EEGs? :thinking_face:
10/17/2017 3:58 AM
alexander.chudinov :
Should I be looking into investing in a board from http://openbci.com?
10/17/2017 4:02 AM
alexander.chudinov :
It seems too simple too just hook up an electrode to an arduino...
10/17/2017 4:55 AM
elsehow :
alexander.chudinov if you�re wanting to roll your own openBCI, i�d actually recommend the texas instruments ADS1299 ADC. this is an informative short read: https://hal.inria.fr/hal-01278245/document
10/17/2017 1:30 PM
benjamindeleener :
Education Changes the World! Check out our latest Q&A with Sayan Faraz and his team from NeurotechUofT student club to see how they push neurotech innovation at the University of Toronto: https://medium.com/neurotechx/q-a-with-neurotechuoft-d46c06f15126
10/17/2017 2:28 PM
alexander.chudinov :
Is reading data from the electrode as easy as or similar to: read(pinNum) in software? What language do most BCI's use?
10/17/2017 2:31 PM
elsehow :
what do you mean by �langauge�? and yes, you�d just be reading a timeseries of voltages (and probably FFTing them)
10/17/2017 2:33 PM
alexander.chudinov :
c, java, etc
10/17/2017 2:33 PM
alexander.chudinov :
Oh, its probably dependent on the microcontroller
10/17/2017 4:00 PM
mrkrause :
Pretty much, though there are bunch of libraries that handle buffering and synching with other devices too. LSL (Lab Streaming Layer) supports a fair number of devices, for example: https://github.com/sccn/labstreaminglayer
10/17/2017 5:05 PM
yannick :
alexander.chudinov it really depends on your goal and project.
10/17/2017 5:07 PM
yannick :
Python, Matlab, Java, C++, etc. all are being used for BCIs, it depends on the project.
You can find different frameworks on the NeuroTechX Awesome BCI List (https://github.com/NeuroTechX/awesome-bci)
10/17/2017 5:09 PM
yannick :
To answer the read(pinNum)& EEG is a time series kind of signal (as pointed out by elsehow). That means that you�ll receive a value (voltage) every x millisecond. Then you�ll have to filter and clean the signal and then probably do some machine learning.
10/17/2017 5:11 PM
yannick :
Some EEG device companies are offering a SDK to do these steps for you and give you some sort of readout of mental state(s) directly, like read(FocusLevel) that would return a value between 0 and 1. (to keep your comparison)
10/17/2017 6:21 PM
fred-simard :
alexander.chudinov Go with Python, this will take you further and faster than most alternatives. Matlab would be my second choice. Later on, if you focus on a specific application {smartphone, embedded, cloud}, you can switch your language for something more aligned with your development platform.
10/17/2017 6:45 PM
alexander.chudinov :
Noted, thank you everybody. 😄 I've heard that Python is great for neural networks. Can anybody confirm this?
10/17/2017 7:27 PM
graeme :
For those keeping score, that's now two years in a row that someone got on stage at the F8 Facebook developer conference, promised revolutionary BCI-based telepathy, and then left the company a couple months later:
10/18/2017 12:37 AM
fred-simard :
graeme makes me think of Silicon Valley the TV show.
10/18/2017 2:25 AM
dano :
Woahhhhh
10/18/2017 2:25 AM
dano :
Active BCI sure is hard, eh
10/18/2017 2:46 AM
yannick :
Who would have thought...
10/18/2017 5:00 AM
graeme :
10/18/2017 1:03 PM
aj :
Me
10/18/2017 4:24 PM
alexdni :
LMFAO Brilliant graeme - that was JUST the piece i was thinking of
10/18/2017 4:38 PM
graeme :
NeuroTechX should name one of our projects "Potato Gun" in honour of that clip."
10/18/2017 4:49 PM
alexdni :
and just figured out what im doing for Halloween this year - Mr.Potato Head sounds like a winner
10/18/2017 5:39 PM
fred-simard :
mrkrause I bet you thought this was going to go unnoticed. Congrats Mat! Extremely interesting and valuable results you got there. http://www.cell.com/current-biology/fulltext/S0960-9822%2817%2931185-5#.Wd_fWd7-Oqs.linkedin
10/18/2017 5:54 PM
mrkrause :
Thanks fred-simard !
Folks interested in (and/or skeptical about) tDCS: Take a look and I'd be happy to answer questions/chat/etc. I am obviously biased, but I think this is one of the clearer looks we have at how tDCS affects the brain.
10/19/2017 1:16 PM
dano :
Haven't read the full text, but this looks awesome! Exactly the kind of study I was hoping someone would do in tDCS. I know you've probably laid this out in methods, but how does this primate tDCS compare to human in terms of skull size/current penetration?
10/19/2017 5:52 PM
yannick :
Hahaha Graeme I had the same clip in mind.
10/19/2017 6:00 PM
mrkrause :
It�s pretty very!
One of the things I disliked about the older animal tCS literature was that it was really hard to compare with human experiments. I guess stimulating through screws put into or through the skull is technically transcranial but&..we�re clearly not going to do that to grandma. So here, we tried to do only stuff that you could do to a healthy human. We used an off-the-shelf stimulator (the StarStim 8 from Neuroelectrics), along with the PI-Stim electrodes they sell. The electrodes went right on the intact scalp with some boring old SignaGel, and so on. So, except for trimming the hair around the electrodes (it got gross otherwise), the setup matches human experiments really well.
We got high-res MRIs and used those to build finite-element models of the heads. Those predict that the field strength is 0.42 V/m for one and 0.68 V/m for the other. That�s pretty comparable to the field strengths measured and estimated in humans (~0.5 to 1 V/m). Apparently, the monkey skull is a bit thinner but denser, so maybe it cancels out. I�d like to go back and verify our model specifically, but it�s tricky to do for tDCS: the recording gear high-pass filters the signal and thus kills any DC offset. It should be possible with tACS (and, in fact that�s already towards the top of my to-do list).
Anyway, thanks for taking a look and I�m glad you liked the paper. We�ve got a bunch more stuff coming down the pipeline. If there�s any sort of parameter that you really want to see tested, I�m happy to try!
10/23/2017 3:59 PM
mrkrause :
Also, here's a free link: https://authors.elsevier.com/a/1VxDg3QW8RiiPl
10/26/2017 7:27 PM
eferdinand :
10/26/2017 8:56 PM
aj :
can't enter without paying $150
10/26/2017 8:56 PM
aj :
😞
10/26/2017 10:41 PM
fred-simard :
sounds like a scam... no offense ferdinand, but no valuable price
10/27/2017 12:44 PM
hassan :
Anyone have worked/is working on using deep learning to classify brain signals (EEG, ECoG, etc.)?
10/27/2017 1:19 PM
yannick :
I�m slowly starting to look into it and I�d love to discuss more about it with other people who�s been trying DL+EEG. ( so I�m gonna follow this thread 🙂 )
10/27/2017 5:52 PM
dano :
Come to <#C4LSDAE95|neurodoro>. We're working on an open source project to use neural nets to classify concentration in a way that can be deployed on a mobile device
10/27/2017 8:12 PM
bciguy :
hassan we (alexandre.barachant, dano et al) had a nice conversation about this a few months back in <#C08PVFWH4|eegdata> you might want to look up. (I assume by DL you are referring to transfer learning specifically).
10/28/2017 1:45 AM
hassan :
Ok cool! I'm interested in ECoG though. bciguy I think that discussion was deleted by slack. It's not exist anymore.
10/28/2017 1:47 AM
hassan :
I will keep an eye on the transfer learning for sure. However, the question is transfer learning from what?
10/28/2017 4:22 AM
bciguy :
that cant be right
10/28/2017 4:26 AM
bciguy :
hassan i think we were discussing autoencoders, but it was a while ago
10/28/2017 4:27 AM
bciguy :
.. ie applying the �automatic feature learning� deep learning paradigm to eeg
10/28/2017 3:59 PM
yannick :
https://github.com/NeuroTechX/ntx_slack_archive/blob/master/eegdata.md
10/28/2017 4:00 PM
yannick :
You can search �autoencoders� and you�ll find that discussion 🙂
10/28/2017 8:23 PM
bciguy :
yannick how cool, is the neurobotx code open? We have a rapidly growing community in the slack - iWinks User Community and as we scale this is seems like the next best thing to paying slacks expensive memberships. Any advice is greatly appreciated 🙏
10/29/2017 5:45 PM
mesca :
I trained a LSTM to classify ERP data accross different stimuli and participants: https://github.com/mesca/turnkey-bci
10/29/2017 6:11 PM
alexandre.barachant :
mesca did you benchmark it against other methods ?
10/29/2017 6:13 PM
mesca :
Yep, against LDA (I reproduced the results from the original paper).
10/29/2017 6:13 PM
mesca :
- .1 AUC
10/29/2017 6:15 PM
alexandre.barachant :
nice
10/29/2017 6:23 PM
dano :
10/29/2017 8:10 PM
hassan :
Thanks yannic and bciguy. I actually found myself was getting involved in that conservation 😂
10/29/2017 8:14 PM
hassan :
I think the best approach to use DL is to augment the neural data we have. I mean in image classification, the images were rotated and shifted and this is how the dataset get augmented. In neural signals...no idea
10/29/2017 8:17 PM
hassan :
mesca Nice work!
10/30/2017 10:26 AM
alexandre.barachant :
hassan: data augmentation : https://hal.inria.fr/hal-01225255/document
10/30/2017 12:55 PM
hassan :
alexandre.barachant Interesting! Thanks
10/30/2017 6:15 PM
ben.tam :
I am not sure if our community is ready for this, as this may slow our growth, but if we each paid a $5 - $10 a month membership fee ... that would unlock a lot of growth opportunities for NTX in general. I mean, people pay more then this for Netflix.
10/30/2017 6:15 PM
ben.tam :
Also.... https://get.slack.help/hc/en-us/articles/206646877-slack-for-education
10/31/2017 5:34 PM
yannick :
Check out the latest Student Club Q&A, this time with NeuroTech@Berkeley.
https://medium.com/neurotechx/q-a-with-neurotech-berkeley-5cc833103ca
Great work guys! (good job on the interview mattherich!)
10/31/2017 8:40 PM
dano :
Idk if we should ever be charging for member ship. Kind of against our mission. I'm also down to, if slack is not working without paying, switch to Matrix or Gitter or something
10/31/2017 10:18 PM
graeme :
This write-up by urish deserves to be in the general channel:
10/31/2017 10:18 PM
graeme :
10/31/2017 10:44 PM
ben.tam :
Yah I agree it would suck to charge people to be part of a community. Alternatively.... this is a whacky idea for sure... but what about using a crowd equity platforms to fund our non profit? We'd structure it so that current execs of NTX maintain full control, but our community gets to buy ownership(non-voting) of NTX.
11/2/2017 11:29 AM
mesca :
11/2/2017 5:39 PM
elizastrickland :
Hi folks! Thought you might be interested in my article about a project from "experimental philosopher" and conceptual artist Jonathon Keats. He worked with neurotech pioneer Jose del Millan at EPFL on an installation called Mental Work, in which participants use a BCI to control shiny pointless machines:
11/2/2017 5:41 PM
alexandre.barachant :
thats the perfect use of BCI 🙂
11/3/2017 6:06 PM
dano :
DevFest presentation from urish showing how to build a web app that reads blinks from Muse.
https://www.youtube.com/watch?time_continue=602&v=F6fI28yFtNM
11/8/2017 2:20 AM
elsehow :
http://news.berkeley.edu/2016/08/03/sprinkling-of-neural-dust-opens-door-to-electroceuticals/ looks like a neural dust prototype
11/8/2017 10:46 AM
pia.michel :
Are there people here working with BCIs for communication in locked-in?
11/8/2017 5:37 PM
d.adamos :
11/8/2017 5:43 PM
aj :
Whoa
11/8/2017 9:55 PM
sydneyneurotechx :
Congrats on the raise graeme !
11/8/2017 10:03 PM
graeme :
Thanks sydneyneurotechx
11/8/2017 10:04 PM
graeme :
We even get to spend a little bit of it on research and hiring some engineers and scientists
11/8/2017 10:04 PM
graeme :
Stay tuned for job announcements
11/8/2017 10:04 PM
bryan_j :
Congrats graeme !
11/8/2017 10:05 PM
graeme :
11/8/2017 10:05 PM
graeme :
Where they have a high-resolution static magnet MRI with head coils in the basement
11/8/2017 10:05 PM
graeme :
Also, we're moving offices (around the corner) into the same building as Synaptive Medical
11/8/2017 10:07 PM
graeme :
(Finally we can break out our ActiChamp again!)
11/8/2017 10:07 PM
sydneyneurotechx :
oh wow!
11/8/2017 10:07 PM
graeme :
And we'll be hosting some brainstorming sessions, hopefully involving NeuroTechTO and NeuroTechHA members, on the design of our new lab space where we'll host some future hack nights
11/8/2017 10:08 PM
sydneyneurotechx :
hectordomorozco ^
11/8/2017 10:11 PM
sydneyneurotechx :
I love the work they do at Synaptive. I'm excited to see that two of my favorite Toronto companies will be so close together.
11/8/2017 10:12 PM
dano :
Ooohh future hack space
11/8/2017 10:12 PM
dano :
You gonna be around tonight, graeme?
11/8/2017 10:12 PM
bryan_j :
We would be happy to help organize and contribute! Thank you!!
11/8/2017 10:12 PM
graeme :
no but make sure you collect ideas for the new lab
11/8/2017 10:13 PM
bryan_j :
Hahahaha woohoo!!
11/8/2017 10:15 PM
dano :
hammock
11/8/2017 10:15 PM
dano :
beer fridge
11/8/2017 10:15 PM
dano :
we'll have a full dossier 😉
11/8/2017 10:54 PM
micslab :
Is there anyone from neurable in our group? Please connect on the <#C44MX942K|_australia> channel if so.
11/8/2017 10:59 PM
yrenard :
adamm_neurable ^^
11/9/2017 5:10 AM
sydneyneurotechx :
11/9/2017 5:18 PM
dano :
^ I think this should be a must-read for all of us. This tech that we're intellectually and professionally interested in will, without a doubt, raise some ethical issues in the future. These are solid recommendations that I personally intend on paying attention to in the years to come.
11/9/2017 5:21 PM
bryan_j :
Was at a panel discussion last night where the main consideration/concern was the ethical implications and regulation of medical technologies.
11/9/2017 5:22 PM
bryan_j :
All this to say it is a concern across the board, and needs to be addressed. Looking forward to reading this tonight!
11/9/2017 5:23 PM
davidevaleriani :
very interesting article, thanks for sharing it sydneyneurotechx
11/9/2017 7:11 PM
hectordomorozco :
graeme Congratulations!!! For sure; we�d love to get involved with this 😃
11/10/2017 5:30 PM
dan :
Hello! If there are any Chicagoans in the group, come on over to <#C42AR7X5X|chicago>!
11/13/2017 3:45 PM
justme :
Hello somebody know something about smart drug like qualia
11/13/2017 3:59 PM
mhough :
pia.michel did anyone get back to you about lockedin?
11/13/2017 4:05 PM
mhough :
justme What are you looking to do?
11/13/2017 5:24 PM
justme :
No sure yet
11/13/2017 5:25 PM
justme :
Try it maybe
11/13/2017 5:26 PM
justme :
They tell it could litteraly change our brain fonction
11/13/2017 5:28 PM
justme :
Combine with norotechx developper it can be realy big
11/13/2017 6:43 PM
omrin :
For the fixed challenge competition, the instructions say "home-made acquisition board" do they mean fully homemade i.e. design and print the PCB in a microfabrication lab or can we buy a raspberry pi and work from there?
11/13/2017 7:32 PM
yannick :
omrin I guess you are talking about the <#C08S3426L|studentclubs> competition, if you don�t mind asking your question there 🙂
11/13/2017 10:22 PM
pia.michel :
no, nothing so far
11/14/2017 3:03 PM
alexander.chudinov :
Are there any uni opportunities regarding neurotech?
11/14/2017 4:03 PM
alexander.chudinov :
Is neuroscience a good start?
11/14/2017 4:05 PM
david.vumbaco :
Always 🙂
11/14/2017 5:50 PM
benbrain :
from someone with an undergrad in neuroscience, i highly recommend taking a minor in computer science if you're going to specialize in neuroscience as technical skills are vital
11/14/2017 6:47 PM
sydneyneurotechx :
Also is there a particular element of neurotech you would like to work on? alexander.chudinov? Do you want to work on the hardware? Would you prefer to do analysis of the data? Or do it all and build a BCI for a specific use case? Having a physics background can also be beneficial if you want to work with MRI/Ultrasound/NIRS
11/14/2017 6:48 PM
sydneyneurotechx :
I'd also look into biomedical engineering. It all depends on what you would like to do. Neurotech is a big field 🙂
11/15/2017 1:59 AM
yannick :
Congratulations to conor (CEO of OpenBCI) for his Forbes 30 under 30 nomination. OpenBCI has allowed a lot of people to learn more about bio-signals like EMG and EEG and has tirelessly been spreading the values of open source and accessibility over the past few years.
https://www.forbes.com/30-under-30/2018/science/#1cba36773eac
11/15/2017 10:29 PM
fred-simard :
Not a game changer, but still: https://www.kickstarter.com/projects/neurotinker/neurobytes-electronic-neuron-simulators
11/15/2017 10:37 PM
alisa :
Hello! My name is Alisa and I am an undergraduate neuroscience student writing on behalf of NeurotechUofT. Would anyone here be very good with EMG?
11/15/2017 10:38 PM
alisa :
We need emergency help regarding a non-NeurotechX competition project.
11/16/2017 2:01 AM
yrenard :
alisa I can probably help with this
11/16/2017 3:25 PM
dano :
Those Neurobytes are actually great fun and have a lot of potential as a neuroscience teaching tool. teon was showing them off at the Mozilla Working Open Workshop
11/16/2017 3:36 PM
dano :
They'd be really great as activities for low-key chapter events
11/16/2017 4:06 PM
bryan_j :
Yupp
11/16/2017 4:06 PM
bryan_j :
Or educational outreach events
11/16/2017 6:40 PM
yannick :
neurojoe ^ people are talking about your NeuroBytes 🙂
11/16/2017 7:01 PM
marion :
For those of you interested in computational neuroscience, Numenta's new theory is out. The companion video is great at summarizing how columns and minicolums in the cortex treat sensory and proprioception information to recognize objects http://www.businesswire.com/news/home/20171115006003/en/Numenta-Publishes-New-Theory-Solve-Mystery-Brain
11/16/2017 10:46 PM
chriscrawford :
Anyone have a reference to changes in the cost of physio sensors over the past decade?
11/17/2017 6:15 PM
bryan_j :
I wonder if we could get some?!
11/17/2017 9:33 PM
hassan :
11/19/2017 2:29 AM
hassan :
Anybody knows if there are characterizations for the ECoG signals from the seizure site or foci? I mean when this site is not in the "seizure state"
11/20/2017 3:55 PM
joyous :
Hi everyone! My name is Joy and I have a budding interest in Neuroscience. I've been focusing primarily on cognitive development and learning, with a focus on how to optimize memory formation. I'm still new to the field and in the early stages of research, so forgive me if this question is a bit basic, but I keep coming across papers in Alzheimer's research that potentially alludes to the idea that there may be some correlation between learning/memory encoding helping to stave off Alzheimers. I was wondering if anyone of you knew if there was any nascent studies or papers about using learning systems to help mitigate Alzheimers degenerative effects, or showing which certain types of learning/education methods have shown to help mitigate the risks of developing Alzheimer�s later in life.
11/21/2017 9:57 AM
pierre :
Here's movement prediction with CNNs and LSTMs on ECoG data: https://arxiv.org/abs/1709.05939
11/21/2017 9:57 AM
pierre :
^ hassan
11/21/2017 10:28 AM
pierre :
joyous Have you seen the work of the Gazzaley lab in this direction? For instance: https://neuroscape.ucsf.edu/neuroscience/#healthy-aging
11/21/2017 10:28 AM
pierre :
mhough might know more, as he is part of the lab
11/21/2017 1:17 PM
hassan :
pierre Awesome! Thank you
11/21/2017 3:46 PM
bryan :
thanks for highlighting this lab. I may reach out in a year or so to see about a post-doc appointment
11/21/2017 7:23 PM
joyous :
I hadn't heard of them, but thank you for sharing! I'll definitely look into them
11/21/2017 8:11 PM
d.adamos :
@channel Hi all, please check on our Neuroinformatics Group EEG Music experiment for TELIA Norway (mobile phone provider) TV/media campaign �Music Freedom�:
https://www.linkedin.com/pulse/unravelling-music-taste-famous-norwegian-artists-decoding-adamos/
11/21/2017 8:15 PM
d.adamos :
@Channel Hi all, please check on our Neuroinformatics GRoup EEG Music experiment for TELIA Norway (mobile phone operator) TV/media campaign �Music Freedom�:
Unravelling the music taste of famous Norwegian artists by decoding their brainwaves
11/22/2017 2:37 PM
bryan_j :
11/22/2017 7:55 PM
nmp256 :
Hey there, not sure if this is the best channel for this question so please let me know if there's a better place for it. I'm a long time meditator interested in how self-compassion can be used to help people with eating disorders. There's already a lot of good research I've found on it, but not much I've found yet that covers how to detect compassion through biometrics (best I can get is looking at calm through heart rate changes). Does anyone here know if EEG can be used to detect anything like compassion and/or acceptance of yourself? Thanks so much!
11/22/2017 9:32 PM
benbrain :
nmp256 The short answer is maybe. It definitely is a complex and nuanced problem and part of it stems from the fact that "compassion" and "acceptance of yourself" are abstract concepts that likely don't have clear or consistent representations in the brain. There isn't a simple or easy way to detect emotional states like these, but that's not to say that going forward there won't be methods developed that can potentially predict the emotional state of a person using EEG data and machine learning. However, as of yet, you would be extremely hard pressed to find a robust and accurate method of detecting those emotional states.
11/22/2017 9:55 PM
aviaminov :
Hello! Has anyone here found interest in EEG hypsarrhythmia, specifically with automatic classification of infantile EEG recording to find such patterns?
11/23/2017 1:12 AM
nmp256 :
benbrain thanks for the response! Anything currently being done with EEG data and machine learning for detecting emotions or these kinds of emotions?
11/23/2017 12:42 PM
fortin.pasc :
benbrain nmp256 I am doing quite a bit of research right now of emotion recognition and to be honest, the vast majority of papers relying on EEG, HRV, GSR, respiration, skin temperature, and video have heavily flawed methodologies (either on purpose or not, but that's another debate) with heavily inflated reported performance. Consequently, I would not trust anyone who claim they can do it with more than extremely coarse resolution.
That being said, although impractical, facial electromyography seems to be the most user-independent and reliable approach for the moment. If you are interested in facial expression of emotional states I would suggest reading a few of Paul Ekman's papers on the universality of emotion expressions and specifically those relating to the Facial Action Coding System (FACS). If you want more and have a bit of free time, I encourage you to read or scan through Charles Darwin's "The Expression of the Emotions in Man and Animals" which is not exactly scientifically sound but initiated a lot of work in this field.
11/23/2017 5:43 PM
nmp256 :
fortin.pasc_ Thanks for the note. I have looked at Ekman's work and was just on the phone with someone using facial microexpressions to recognize emotions to aid therapy for people with PTSD. Any pointers to the cutting edge papers that speak to these findings? Have you published or written a literature review or anything of that sort?
11/23/2017 5:46 PM
nmp256 :
fortin.pasc Also, for the things you've seen do they try and find correlations between self reported emotional states using known validated measurement tools and these other sensors?
11/24/2017 3:53 PM
fortin.pasc :
I have not published or written a literature review on this topic yet, and I do not have a recent EMG paper handy.
For your second question, yes they did/do/will do, and that is until someone finds a way of collecting ground truth for emotional states. Currently, for coming with new emotion recognition methods and validating them, the common approach is to run a lab experiment where you do emotion elicitation using different datasets containing pictures, videos, emotional stimuli. They sample the subject's affective state using tools from the psychology literature, and then use the results of these tools as ground truth for their training/correlation.
11/26/2017 3:09 PM
gerrie :
nmp256 I was reminded that one of the Emotiv applications claimed to be able to classify emotional states. They were very rough (Happy, sad, calm, stressed - if I recall correctly). Nothing as abstract as "acceptance of yourself". Also, if my memory serves me, we were not blown away by the accuracy or repeatability of the classification. Having said that, it may be interesting to explore the basis for their classification.