BAMF Lung and FDG-Avid Tumor

[jithenece]

Pretrained model for 3D semantic image segmentation of the FDG-avid lesions from PT/CT scans

[jithenece]

sample:
  idc_version: Version 5: Updated 2020/12/22
  data:
  - SeriesInstanceUID: 1.3.6.1.4.1.14519.5.2.1.6655.2359.184691786899278131062806456462
    aws_url: s3://idc-open-data/aafd2a22-6236-4eff-b541-1dfd8de923e8/*
    path: case_study1/ct
  - SeriesInstanceUID: 1.3.6.1.4.1.14519.5.2.1.6655.2359.168809572664004911257595338967
    aws_url: s3://idc-open-data/c6ff3b5f-eb4d-4a08-8e6b-3004b2ebbc0f/*
    path: case_study1/pt

reference:
  url: https://drive.google.com/file/d/1jcujIoSvYG0Owps8nZVBBbjSGnCk7Wdx/view?usp=sharing

[jithenece]

screenshots.zip

Added screenshots

[jithenece]

@LennyN95 could someone review this so that i can do similar fixes for other PRs and submit?

[LennyN95]

Overall great implementation. I've made some points on the usage of TS as part of this model that I'd like to discuss briefly before moving forward.

[LennyN95]

Looks good! Some minor suggestions below.

[jithenece]

/review

@LennyN95 Please review and let me know for more changes.

[jithenece]

/test

sample:
  idc_version: Version 5: Updated 2020/12/22
  data:
  - SeriesInstanceUID: 1.3.6.1.4.1.14519.5.2.1.6655.2359.184691786899278131062806456462
    aws_url: s3://idc-open-data/aafd2a22-6236-4eff-b541-1dfd8de923e8/*
    path: 'case_study1/ct'
  - SeriesInstanceUID: 1.3.6.1.4.1.14519.5.2.1.6655.2359.168809572664004911257595338967
    aws_url: s3://idc-open-data/c6ff3b5f-eb4d-4a08-8e6b-3004b2ebbc0f/*
    path: 'case_study1/pt'

reference:
  url: https://drive.google.com/file/d/1MGe1dR22GF-oF7BwQ4bE3Oq-9KBnohku

[jithenece]

/test

sample:
  idc_version: "Data Release 5.0 December 22, 2020"
  data:
  - SeriesInstanceUID: 1.3.6.1.4.1.14519.5.2.1.6655.2359.184691786899278131062806456462
    aws_url: s3://idc-open-data/aafd2a22-6236-4eff-b541-1dfd8de923e8/*
    path: 'case_study1/ct'
  - SeriesInstanceUID: 1.3.6.1.4.1.14519.5.2.1.6655.2359.168809572664004911257595338967
    aws_url: s3://idc-open-data/c6ff3b5f-eb4d-4a08-8e6b-3004b2ebbc0f/*
    path: 'case_study1/pt'

reference:
  url: https://drive.google.com/file/d/1MGe1dR22GF-oF7BwQ4bE3Oq-9KBnohku/view?usp=sharing

[jithenece]

/test

attaching segmentation
output.zip

sample:
  idc_version: 15.0
  data:
  - SeriesInstanceUID: 1.3.6.1.4.1.14519.5.2.1.4334.1501.322501480363502116767369539775
    aws_url: s3://idc-open-data/7d19e1ee-f2c9-4158-a6e2-d093468e393b/*
    path: case1/ct
  - SeriesInstanceUID: 1.3.6.1.4.1.14519.5.2.1.4334.1501.329094085186214039017114090511
    aws_url: s3://idc-open-data/c39ec1fc-f5d3-4685-9077-dafd79bc5970/*
    path: case1/pt

reference:
  url: https://github.com/user-attachments/files/16797384/output.zip

---

Test Results (24.08.29_15.15.30_AkoraywFNq)

id: 02819301-0b3b-4324-9723-77207b8708d6
date: '2024-08-29 16:21:07'
checked_files:
- file: bamf_pet_ct_lung_tumor.seg.dcm
  path: /app/test/src/case1/bamf_pet_ct_lung_tumor.seg.dcm
  checks:
  - checker: DicomsegContentCheck
    notes:
    - label: Segment Count
      description: The number of segments identified in the inspected dicomseg file.
      info: 2
    findings:
    - label: Dice Score Difference
      description: Dice score between reference and test image
      subpath: 'segment #1'
      info: 0.975907953098239
summary:
  files_missing: 0
  files_extra: 0
  checks:
    DicomsegContentCheck:
      files: 1
      findings:
        Dice Score Difference: 1
conclusion: false

[jithenece]

/test

attaching segmentation
output.zip

sample:
  idc_version: 15.0
  data:
  - SeriesInstanceUID: 1.3.6.1.4.1.14519.5.2.1.4334.1501.322501480363502116767369539775
    aws_url: s3://idc-open-data/7d19e1ee-f2c9-4158-a6e2-d093468e393b/*
    path: case1/ct
  - SeriesInstanceUID: 1.3.6.1.4.1.14519.5.2.1.4334.1501.329094085186214039017114090511
    aws_url: s3://idc-open-data/c39ec1fc-f5d3-4685-9077-dafd79bc5970/*
    path: case1/pt

reference:
  url: https://github.com/user-attachments/files/16927593/output.zip

[LennyN95]

@jithenece the test results suggest there is a quite high difference (DiceScore), could you elaborate on this?

Please note, we updated our base image. All mhub dependencies are now installed in a virtual environment under /app/.venv running Python 3.11. Python, virtual environment and dependencies are now managed with uv. If required, you can create custom virtual environments, e.g., uv venv -p 3.8 .venv38 and use uv pip install -p .venv38 packge-name to install dependencies and uv run -p .venv3.8 python script.py to run a python script.

We also simplified our test routine. Sample and reference data now have to be uploaded to Zenodo and provided in a mhub.tom file at the project root. The process how to create and provide these sample data is explained in the updated testing phase article of our documentation. Under doi.org/10.5281/zenodo.13785615 we provide sample data as a reference.