Answers are grounded in this paper and cite their sources.
Context.— Pathology departments transitioning to a digital workflow are encouraged to adopt the digital imaging and communications in medicine (DICOM) format for routine diagnosis, leading to a substantial increase in histopathological image-based research projects using this format. Nevertheless, DICOM for whole-slide images (WSIs) remains an emerging file type, with few tools currently available for de-identification that do not require coding skills. Objective.— To develop an easy-to-use, no-code tool for de-identifying DICOM WSIs and providing feedback on the de-identification process. Design.— An open-source and Python-based solution, with detailed explanations, is proposed to de-identify DICOM WSI files, relying on the Pydicom package. The de-identification process removes the label image and the macro/overview image, and modifies DICOM metadata in accordance with the DICOM Basic Application Level Confidentiality Profile. The native Python script was converted with PyInstaller into executables compatible with Windows and macOS. Tests for residual protected health information were performed using 2 distinct open-source solutions. Results.— The solution was tested on 73 DICOM slides coming from 5 major vendors (Leica, 3DHistech, Roche, Hamamatsu, and Olympus/Evident), encompassing brightfield and fluorescence modalities, single- and multi-plane acquisitions, and single- and multi-plex images. Average de-identification time did not exceed 1 or 2 s/GB. The executables can be used without any coding expertise, while providing access to the remaining metadata for compliance verification. Conclusions.— An efficient solution for de-identifying DICOM WSI is presented, addressing a practical need among pathologists who lack programming skills but wish to use digitized slides for research and educational purposes.
No takes yet. Share an insight, caveat, or question.
Bertrand Chauveau (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: