Pydicom

If you're building anything that touches medical imaging, this is your entry point to DICOM files. It handles the full workflow: reading CT/MRI/X-ray data, extracting pixel arrays for analysis, anonymizing patient info before sharing datasets, and converting to standard formats like PNG. The skill covers practical gotchas like applying windowing for proper CT display, handling multi-frame sequences, and dealing with compressed transfers that need extra codecs. DICOM is notoriously finicky with its metadata tags and transfer syntaxes, so having clear examples for both reading existing files and writing new ones from numpy arrays saves real debugging time. Good for radiology pipelines, research datasets, or PACS integration work.

skills.sh installs: 2K
GitHub stars: 56.6K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add k-dense-ai/scientific-agent-skills --skill pydicom --agent claude-code

Installs into .claude/skills of the current project.

Files

pydicom

Use pydicom for DICOM dataset I/O and pixel processing. Version 3.0.2 is the current stable release reviewed here. It fixes CVE-2026-32711, a crafted DICOMDIR path-traversal issue. pydicom 3.0.2 declares Python >=3.10; its bundled DICOM dictionary is 2024c; the current DICOM Standard reviewed here is 2026d. No PACS, DIMSE, DICOMweb endpoint, or authenticated service is exercised by this local-file skill.