Automated and Interpretable Detection of Hippocampal Sclerosis
AID-HS extracts hippocampal volume- and surface-based features from T1w scans using HippUnfold, and provides an in-depth characterisation of hippocampal abnormalities as well as the automated detection and lateralisation of hippocampal sclerosis (HS). For more details please read our manuscript)
Note:
- You will need the following demographic information (age at scan & sex) to run AID-HS on your patient's T1 MRI scan.
- AID-HS has been developed on T1w scans acquired at 3T. It has not yet been thoroughly evaluated on 1.5T and 7T data
Pipeline overview:
The AID-HS software is intended for research purposes only and has not been reviewed or approved by the Medicines and Healthcare products Regulatory Agency (MHRA), European Medicine Agency (EMA) or by any other agency. Any clinical application of the software is at the sole risk of the party engaged in such application. There is no warranty of any kind that the software will produce useful results in any way. Use of the software is at the recipient's own risk.
You can install and use the AID-HS pipeline with :
- docker container recommended for easy installation of the pipeline as all the prerequisite packages are already embeded into the container. Note: Dockers do not work on High Performance Computing (HCP) systems.
- singularity container (COMING SOON) - to run a container on High Performance Computing (HCP) systems.
- native installation: Not supported
Once installed you will be able to use the AID-HS pipeline on your data following the steps:
- Prepare your data : guidelines
- (OPTIONAL) Compute the harmonisation parameters : guidelines
- Run the prediction pipeline: guidelines
- Interpret the results: guidelines
What is the harmonisation process ?
Features extracted from MRI scans from different MRI scanners have systematic differences between them. To remove scanner related biases we recommend harmonising your MRI data to the MRI data that was used in the AID-HS manuscript. This harmonisation is required for each MRI scanner / T1 sequence you are using.
Notes:
- This step needs to be run only once, and requires data from at least 20 subjects acquired on the same scanner with the same T1 sequence and demographic information (e.g age and sex). See harmonisation instructions for more details.
- The AID-HS pipeline can also be run without harmonisation with no drop in performances. However, the characterisation of the hippocampal features compared to the normative growth curves will not be interpretable.
If you are using the AID-HS tool or part of the code, please cite:
An overview of the notebooks that we used to create the figures can be found here.
Mathilde Ripart, PhD student, UCL
[email protected]