design-study

Analyzes radiology AI study designs for validity risks and improvements.

243|60|Updated Apr 6, 2026
One-click install
npx skills add https://github.com/Aperivue/medsci-skills --skill design-study
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: design-study
Source: https://github.com/Aperivue/medsci-skills/tree/main/skills/design-study
Command: npx skills add https://github.com/Aperivue/medsci-skills --skill design-study

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill provides an early, structured validity diagnostic to ensure a study question is answerable, analysis plan is coherent, and reporting expectations are aligned before drafting or data work begins.

Core Features & Use Cases

  • Structured reconstruction: Reconstructs the study protocol to confirm population, inputs, timing, and outcomes align with the research question.
  • Leakage and validation checks: Identifies potential information leakage, unit-of-analysis mismatches, and validation strategy feasibility.
  • Clinical framing guidance: Assesses whether the comparator, endpoint definitions, and claimed benefits are clinically meaningful.
  • Workflow guidance: Suggests path changes and handoff routes (analyze-stats, check-reporting, self-review, or write-paper) based on the findings.

Quick Start

Provide the study protocol or draft describing the population, inputs, and endpoints to receive a validity-focused review of design risks and recommended fixes.

Frequently Asked Questions about design-study

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I check for data leakage in radiology AI study design?

A validity diagnostic checks whether inclusion criteria, timing, endpoints, and reference standards align with the intended clinical use. It ensures the research question is answerable and the analysis plan is coherent before data work or manuscript drafting begins.

How do I validate study endpoints and reference standards for radiology AI research?

You provide the study protocol or draft describing the population, inputs, and endpoints. The skill then analyzes this input to output a structured review detailing major validity risks, recommended fixes, and suggested next steps.

What is the best way to identify unit-of-analysis mismatches in medical imaging AI studies?

It provides a structured validity diagnostic that reviews inclusion criteria, timing, endpoints, and reference standards. This confirms the population, inputs, and outcomes align with the research question before drafting or data work begins.

Can I use a study design review to prepare a radiology AI manuscript?

It suggests path changes and handoff routes, directing users to analyze-stats, check-reporting, self-review, or write-paper workflows. These recommendations are based on the specific leakage risks and validation issues found during the protocol review.