Interoperable slide microscopy viewer and annotation tool for imaging data science and computational pathology

Abstract

The exchange of large and complex slide microscopy imaging data in biomedical research and pathology practice is impeded by a lack of data standardization and interoperability, which is detrimental to the reproducibility of scientific findings and clinical integration of technological innovations. We introduce Slim, an open-source, web-based slide microscopy viewer that implements the internationally accepted Digital Imaging and Communications in Medicine (DICOM) standard to achieve interoperability with a multitude of existing medical imaging systems. We showcase the capabilities of Slim as the slide microscopy viewer of the NCI Imaging Data Commons and demonstrate how the viewer enables interactive visualization of traditional brightfield microscopy and highly-multiplexed immunofluorescence microscopy images from The Cancer Genome Atlas and Human Tissue Atlas Network, respectively, using standard DICOMweb services. We further show how Slim enables the collection of standardized image annotations for the development or validation of machine learning models and the visual interpretation of model inference results in the form of segmentation masks, spatial heat maps, or image-derived measurements.

NCI Cancer Research Data Commons: Resources to Share Key Cancer Data

“…Tools available to IDC users include: (i) IDC-maintained tools (ii) IDC search portal integrated with image visualization tools (iii) Customized instance of the Open Health Imaging Foundation (OHIF) radiology viewer for visualization of radiologic modalities images and image-derived data ( 18 ) (iv) Customized instance of the Slim microscopy viewer ( 19 ) for visualization of digital pathology and microscopy images, and related image-derived data (v) OpenSlide ( 20 ) DICOM supports reading DICOM Slide Microscopy format within the widely used library providing a common interface to access a variety of image formats (vi) Bio-Formats ( 21 ) DICOM supports for reading and writing DICOM Slide Microscopy format (vii) Tools for harmonization of research and vendor-specific formats (i.e., TIFF, SVS, NRRD, NIFTI) into DICOM (viii) Collaborative tools (ix) Google Healthcare API and BigQuery: metadata accompanying IDC data, as available in DICOM files, is automatically extracted, versioned and is made available for searching using Standard Query Language (SQL) queries (x) Google Cloud Platform (GCP): colocation of data within Google Cloud Platform enables scalable access to a variety of components within GCP, enabling the use of popular desktop applications, such as 3D Slicer ( 22 ), or batch image analysis tools, such as automatic segmentation using nnU-Net family of algorithms ( 23 ) (xi) Other tools include Google Data Studio, used to build custom dashboards for data exploration, and Google Colab to streamline prototyping and dissemination of analysis workflows (xii) MHub ( https://mhub.ai ): a repository of self-contained deep-learning models trained for a wide variety of applications in the medical and medical imaging domain. AI tools in MHub are curated with standardization and integration with IDC in mind, to simplify application of those tools to IDC data and integration of the analysis results back into IDC …”

“…(iv) Customized instance of the Slim microscopy viewer ( 19 ) for visualization of digital pathology and microscopy images, and related image-derived data…”

Digital Pathology: A Comprehensive Review of Open-Source Histological Segmentation Software

“…It is an essential process for the analysis reproducibility. In fact, standardization is relevant for preserving the quality and integrity of image data, minimizing the impact of noise, artifacts and errors [142].…”

Thirty Years of the DICOM Standard

“…Recently, DICOM has added a protocol called DICOMweb built on top of the Hy-perText Transfer Protocol (HTTP) for using services via the web [17][18][19][20]. DICOMweb enables query, retrieval, storage, and worklist services.…”