How can I visualize the imaging data before downloading it?

The skill can generate unique viewer URLs for specific Study or Series Instance UIDs, allowing you to inspect medical images directly in your browser using the IDC viewer.

Do I need an account to download data from the Imaging Data Commons?

No, this skill provides access to public datasets that do not require authentication or a specialized cloud provider account for standard queries and downloads.

What imaging modalities are supported by this skill?

The skill supports various modalities including radiology (CT, MR, PET) and digital pathology slide microscopy (SM), along with derived objects like segmentations and annotations.

Are there costs associated with downloading large datasets?

No, IDC data is hosted on public cloud buckets with no egress fees, making it ideal for large-scale data collection for research or AI training.

Can I use the data retrieved through this skill for commercial research?

Most IDC data is provided under CC-BY licenses, but some collections are restricted to non-commercial use (CC-NC). The skill includes metadata to check specific licenses for each dataset.

Imaging Data Commons Access

Name: Imaging Data Commons Access
Author: henriquescastilho

byhenriquescastilho

•

数据科学与机器学习

Accesses and manages large-scale public cancer imaging datasets from the NCI Imaging Data Commons using the idc-index tool.

This skill enables researchers and developers to seamlessly query, visualize, and download public cancer imaging data (CT, MR, PET, and pathology) from the National Cancer Institute's Imaging Data Commons (IDC). By leveraging the idc-index Python package, users can filter massive datasets by metadata, anatomical site, or modality without requiring authentication. It simplifies the discovery of high-quality DICOM data for AI model training, clinical research, and medical imaging analysis, providing direct access to metadata tables, clinical data, and public cloud storage locations.

主要功能

01Download DICOM images directly from public cloud storage (S3/GCS) without authentication

02Query extensive metadata using SQL for radiology and pathology datasets

03Access structured clinical data and join it with imaging metadata for complex analysis

04Visualize medical imaging series in-browser via generated viewer URLs

051 GitHub stars

06Filter data by modality, cancer type, anatomical site, and expert-curated analysis results

使用场景

01Building large-scale AI training datasets for cancer detection using public DICOM images

02Conducting clinical research by joining patient metadata with radiology and pathology scans

03Validating AI-generated segmentations against expert-curated ground truth from IDC collections

主要功能

01Download DICOM images directly from public cloud storage (S3/GCS) without authentication

02Query extensive metadata using SQL for radiology and pathology datasets

03Access structured clinical data and join it with imaging metadata for complex analysis

04Visualize medical imaging series in-browser via generated viewer URLs

051 GitHub stars

06Filter data by modality, cancer type, anatomical site, and expert-curated analysis results

使用场景

01Building large-scale AI training datasets for cancer detection using public DICOM images

02Conducting clinical research by joining patient metadata with radiology and pathology scans

03Validating AI-generated segmentations against expert-curated ground truth from IDC collections