NA-MIC Project Weeks

Back to Projects List

Optimize DICOMweb access to IDC data

Key Investigators

• Daniela Schacherer (Fraunhofer MEVIS)
• David Clunie (PixelMed, USA)
• Steve Pieper (Isomics Inc., USA)
• Andrey Fedorov (BWH, USA)
• Joël Spaltenstein (Agora Care SA, Switzerland)

Project Description

As the DICOM standard is increasingly used in digital pathology imaging, conversion of available datasets from proprietary formats into DICOM format can make the data more FAIR and improve transparency and reproducibility of research conducted with these data. For this reason, the NCI Imaging Data Commons (IDC) hosts all its data in DICOM format.

DICOMweb™ is a set of RESTful services that allows search, access and storage of DICOM objects via network. IDC data are also available (currently outdated, but soon to be updated) from a DICOM store. This project aims to specifically try to optimize access to digital pathology IDC data via DICOMweb capable libraries (wsidicom, dicomslide, ez-wsi-dicomweb) based on prior works on that issue.

Objective

1. Objective A: Have a code snippet for randomly and efficiently selecting tiles of arbitrary size from the IDC-DICOM store.
2. Objective B (if time permits): Have an efficiency comparison for different libraries used for DICOMweb access.

Approach and Plan

Objective A:

1. Summarize current state.
2. Implement DICOMweb access (problems have priorily been observed wrt authorization).

Objective B:

1. Create scripts for efficiency comparison.
2. Summarize results.

Progress and Next Steps

Objective A:

1. Current state summarized in document (will be made public following PW).
2. Created scripts for access with dicomslide, wsidicom, ez-wsi-dicom.

Objective B:

1. Time didn’t permit further efficiency comparison. But based on previous results and discussions, it became clear that ez-wsi-dicom is currently the most efficient tool probably due to the use of multiple ways of parallelizing requests and caching.
2. The main downside of ez-wsi-dicom is that it only works with Google DICOMweb stores.

Illustrations

Conceptual overview of DICOMweb. Taken from: https://www.dicomstandard.org/using/dicomweb/capabilities.

Background and References

Background reading:

• https://www.dicomstandard.org/using/dicomweb

Further resources:

• Related earlier project from Project Week 40: Current state of DICOMweb for pathology.
• Extensive tutorial notebook of ez-wsi_dicomweb: Getting Started with EZ-WSI DICOMweb