medical-imaging-review_skill

Overview

This skill produces comprehensive literature reviews for medical imaging AI research, following a systematic 7-phase workflow and domain-specific templates. It is designed to create structured survey papers, systematic reviews, and critical literature analyses across modalities such as CT, MRI, X-ray, ultrasound, and pathology imaging.

How this skill works

The skill guides you through staged execution: project initialization, targeted literature search across sources (preprints and peer-reviewed), structured method synthesis, quantitative comparisons, and manuscript drafting with required elements (key points, comparison tables, metrics, and figure placeholders). It enforces hedging language, citation support for claims, and explicit limitations for every method section.

When to use it

• Writing a survey or state-of-the-art paper on segmentation, detection, or classification in medical imaging.
• Preparing a systematic review or meta-analysis that requires standardized evaluation metrics and dataset summaries.
• Creating a critical methods comparison for a grant, thesis chapter, or conference tutorial.
• Mapping clinical translation and regulatory readiness of imaging AI products.
• Generating manuscript-ready drafts with figures, tables, and organized references for submission.

Best practices

• Adopt hedging language and avoid absolute statements; support each claim with citations.
• Include a 3–5 bullet Key Points summary and a limitations paragraph for every method category.
• Use standardized performance metrics (e.g., Dice, HD95) and report values with dataset context.
• Provide comparison tables per major section and clear figure placeholders with descriptive captions.
• Search multiple sources (arXiv for recent methods, PubMed for clinical validation) and collate references centrally.

Example use cases

• A systematic review comparing deep learning segmentation methods for brain MRI with Dice/HD95 meta-analysis.
• A survey paper summarizing detection and classification advances in chest X-ray AI, including public dataset table.
• A literature analysis highlighting clinical validation and regulatory steps for AI products in pathology imaging.
• A methods-focused chapter that organizes architectures, training strategies, and failure modes across ultrasound tasks.