Medical Imaging Review

If you're writing a medical imaging AI review paper for journal submission, this skill enforces the discipline that prevents you from shipping a draft full of hallucinated citations and placeholder numbers. It walks you through a six-phase workflow that makes you verify every claim before you write it, not after. The methodology is opinionated: no flat method catalogs, no AI-detector hedge phrases, no vendor names in body text, and verdict sentences only where you have evidence to back them up. Built directly from the failure modes of a coronary CTA review that shipped with 17 broken DOIs and fabricated architecture details. It's overkill for internal lit reviews, but if the goal is to pass peer review on factual grounds, it's the right amount of paranoia.

skills.sh installs: 695
GitHub stars: 885
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add luwill/research-skills --skill medical-imaging-review --agent claude-code

Installs into .claude/skills of the current project.

Files

Medical Imaging AI Review

Build traceable review artifacts whose claims can be audited back to first sources. Report the actual project state and unresolved human gates. Do not promise editorial outcomes or describe an incomplete project as a completed review.

Scope and hard boundary

Support exactly four route tokens across three methodological families:

User intent	Route token	Route
Interpret a field	narrative	Narrative review
Compare method families	method-survey	Method survey
Map available evidence and research gaps	scoping	Scoping review
Answer a focused question without statistical pooling	systematic	Systematic review with qualitative synthesis