Read DICOM metadata
Discover files recursively and parse metadata without loading pixel data.
DICOM metadata quality control
dicomqc
Independent metadata audit for DICOM research releases.
A read-only command-line framework that evaluates candidate DICOM data after external pseudonymization or de-identification. It turns explicit policy checks into reviewable JSON, CSV, and MultiQC evidence.
Current audit surface
A focused metadata control point
dicomqc evaluates the output of a transformation workflow. Keeping that check independent makes release criteria visible, testable, and reproducible across tools and data providers.
Apply explicit rules
Identify direct PHI fields, pseudonym-pattern failures, and private tags.
Produce review evidence
Write JSON, CSV, and MultiQC outputs without reporting raw DICOM values.
Drive pipeline decisions
Use stable exit codes and structured findings in repeatable release workflows.
Scope boundary
Evidence, not transformation
Version 0.1 does not pseudonymize or modify DICOM files, inspect pixels or facial features, or certify DICOM PS3.15, BIDS, HIPAA, or GDPR compliance. Findings support technical and institutional review; they do not replace it.
Documentation
Follow the task at hand
Install dicomqcInstall the PyPI release, optional MultiQC support, or a source checkout.Run an auditGenerate the demo, scan a directory, and interpret the result.Plan remediationApply findings with an external DICOM transformation tool and audit again.Review the architectureTrace metadata through discovery, policy evaluation, and report generation.Compare prior workUnderstand how dicomqc relates to existing de-identification software.