NA-MIC Project Weeks

Back to Projects List

Training AI algorithms on IDC data

Key Investigators

• Cosmin Ciausu (Brigham and Women's Hospital, USA)
• Andrey Fedorov (Brigham and Women's Hospital, USA)

Presenter location: In-person

Project Description

Imaging Data Commons provides publicly available cancer imaging data.

Previous works(IDC Prostate segmentation) (NLST-Body Part Regression) demonstrated through several use cases inference and analysis of AI algorithms on IDC data. Downloading IDC data, conversion between file imaging standards, cloud environment setup and imaging pre-processing steps were studied through these inference and analysis use cases.

During this project week, our goal is to develop use cases of training AI algorithms on IDC data. We welcome any Project Week participants that are interested in leveraging IDC data for training AI algorithms(or evaluation) to collaborate with us!

Objective

1. Leverage IDC data for SOTA segmentation algorithm(nnUNet, MONAI)
2. Collaborate with other members to study the feasibility of using IDC data for training AI algorithms.

Approach and Plan

1. Using nnUNet segmentation framework for prostate segmentation on IDC data(Prostatex/QIN collection) for training purposes.
2. Expand AI training use cases beyond SOTA algorithms.

Progress and Next Steps

1. Leverage information gained by applying inference using nnUNet prostate segmentation on several prostate imaging collections, for training pipelines.
2. Creation of whole prostate IDC training cohort: 45 T2W MRI scans and corresponding expert whole prostate annotations were used.
3. Creation of Google Colab use case showing how to build this cohort and begin a nnUNet training experiment.

Illustrations

Background and References

• Google Colab use case for IDC training cohort and nnUNet training.
• PW37_2022_Virtual – nnUnet - Prostate segmentation on Imaging Data Commons(IDC) data
• AI Imaging analysis on IDC data