NA-MIC Project Weeks

Back to Projects List

Evaluation of interoperability between DICOM WSI Infrastructure components within Kaapana

Key Investigators

• Maximilian Fischer (German Cancer Research Center, Heidelberg)
• Marco Nolden (German Cancer Research Center, Heidelberg)
• Klaus Maier-Hein (German Cancer Research Center, Heidelberg)
• Daniela Schacherer (Fraunhofer MEVIS, Germany)
• Andrey Fedorov (BWH, USA)
• David Clunie (Pixelmed Publishing, USA)

Project Description

Kaapana is an open-source research platform for radiological imaging, featuring a range of infrastructure components. While some of these components natively support DICOM WSI (Whole Slide Imaging), others do not, necessitating the integration of alternative solutions to enable the processing of DICOM WSI data within the platform. The currently integrated components were selected based on their ability to provide fundamental functionality for DICOM WSI data processing, with a focus on being open-source and ensuring high interoperability.

Since the initial integration, newer tools and technologies have been developed that may better meet Kaapana’s requirements. The objective of this project is to reassess the existing components and evaluate whether additional or alternative components should be incorporated. This evaluation aims to enhance the overall processing capabilities for WSI data within Kaapana and further improve the interoperability of the system.

Some components, such as the integrated PACS (dcm4che), are predefined as they form the foundation of the entire platform. Other components, such as the SlimViewer, were integrated on top because it was one of the first viewers capable of connecting to a DICOMweb PACS system.

Goal of this project is to evaluate all exchangeable components to ensure optimal support of DICOM WSI and interoperability with other systems (e.g. for generated analysis results of WSI within the platform).

Objective

1. Identification of available alternatives to visualize DICOM WSI files
2. Test integration possibilities in Kaapana
3. Document results

Approach and Plan

1. Bind other viewing components like QuPath to the data storage in Kaapana
2. Run dl-based analysis of DICOM WSI data with different reading libraries (openslide vs. wsidicom)
3. Visualize exported results (e.g. heatmaps as StructuredReport) with external DICOM WSI viewers.

Progress and Next Steps

1. Describe specific steps you have actually done.

Illustrations

No response

Background and References

• Conversion routines for SR, ANN and fractional SEG developed by Chris Bridge / IDC: https://github.com/ImagingDataCommons/idc-sm-annotations-conversion/tree/main for the dataset available in https://zenodo.org/records/14041167
• Tutorials created by Daniela in IDC for ANN and SR: https://github.com/ImagingDataCommons/IDC-Tutorials/tree/master/notebooks/pathomics (ANN) and https://github.com/ImagingDataCommons/IDC-Tutorials/tree/master/notebooks/collections_demos/rms_mutation_prediction (SR)
• Support of external PACS (eg, Googe Healthcare DICOM stores) in Kaapana: https://codebase.helmholtz.cloud/kaapana/kaapana/-/merge_requests/415, https://codebase.helmholtz.cloud/kaapana/kaapana/-/issues/1191

No response