How does it ensure medical code accuracy?

It integrates with the UMLS MCP server to perform real-time searches, concept lookups, and crosswalks between different code systems like SNOMED-CT and ICD-10CM.

What is the primary purpose of the Synthea skill?

The skill automates the generation of synthetic clinical patient data in FHIR format based on PheKB phenotype definitions, typically used for testing medical software and LLM query accuracy.

Can it handle complex clinical logic?

Yes, it is specifically designed to handle multi-path phenotypes where patients may be identified through different combinations of labs, medications, and diagnosis codes.

What infrastructure is required to run this skill?

It requires a Synthea source build, a Java environment (JDK 17+), and access to a FHIR server (like HAPI or Azure FHIR) if you intend to load the generated data.

Synthea FHIR Test Data Generator

Name: Synthea FHIR Test Data Generator
Author: feordin

byfeordin

0•

데이터 과학 및 ML

Generates high-fidelity synthetic FHIR patient data from PheKB clinical phenotype definitions using Synthea and UMLS.

This skill automates the creation of specialized clinical test data by translating PheKB phenotype definitions into executable Synthea Generic Module Framework (GMF) modules. It leverages UMLS integration to validate and enrich medical codes—including ICD-10, SNOMED-CT, LOINC, and RxNorm—ensuring that synthetic patients adhere to realistic clinical criteria. The skill handles the entire data lifecycle, from simulating multi-path clinical decision trees and generating patient bundles to loading the results into FHIR servers for application testing and LLM query evaluation.

주요 기능

01End-to-end pipeline: create module, generate data, and load to FHIR server

02Automated Synthea module generation from PheKB phenotype documentation

030 GitHub stars

04Environment-aware execution handling Java/Gradle paths in bash or CMD

05Simulation of complex multi-path clinical identification algorithms

06UMLS-backed medical code validation and crosswalk (ICD/SNOMED/RxNorm/LOINC)

사용 사례

01Creating high-fidelity clinical datasets for healthcare application testing

02Benchmarking LLM accuracy for natural language to FHIR query translation

03Simulating specific patient cohorts for medical software development and QA

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add feordin/llm-fhir-query-eval synthea

For use in Claude.ai and ChatGPT

주요 기능

01End-to-end pipeline: create module, generate data, and load to FHIR server

02Automated Synthea module generation from PheKB phenotype documentation

030 GitHub stars

04Environment-aware execution handling Java/Gradle paths in bash or CMD

05Simulation of complex multi-path clinical identification algorithms

06UMLS-backed medical code validation and crosswalk (ICD/SNOMED/RxNorm/LOINC)

사용 사례

01Creating high-fidelity clinical datasets for healthcare application testing

02Benchmarking LLM accuracy for natural language to FHIR query translation

03Simulating specific patient cohorts for medical software development and QA

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add feordin/llm-fhir-query-eval synthea

For use in Claude.ai and ChatGPT