decision-curve-analysis

Official

Assess AI model clinical utility with DCA.

AuthorEvidenceOS
Version1.0.0
Installs0

System Documentation

What problem does it solve?

This Skill addresses the critical gap in evaluating health AI models by determining if they lead to better clinical decisions, beyond just discrimination or calibration.

Core Features & Use Cases

  • Net Benefit Calculation: Quantifies the clinical utility of a model across various risk thresholds.
  • Decision Curve Visualization: Plots net benefit against threshold probability to identify where a model adds value compared to standard strategies.
  • Use Case: Evaluating a new AI model that predicts sepsis risk. DCA helps determine if using the model to guide treatment decisions (e.g., initiating antibiotics) is more beneficial than treating all patients or treating none, across different patient risk levels.

Quick Start

Run the decision-curve-analysis skill to generate a DCA plot for your model predictions and true outcomes.

Dependency Matrix

Required Modules

dcurvespandasnumpy

Components

scriptsreferences

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: decision-curve-analysis
Download link: https://github.com/EvidenceOS/awesome-health-ai-skills/archive/main.zip#decision-curve-analysis

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 223,000+ vetted skills library on demand.