Does this skill support the MIMIC-IV dataset?

Yes, PyHealth provides standardized interfaces for MIMIC-IV, as well as MIMIC-III, eICU, OMOP, and several physiological datasets like SleepEDF.

Can PyHealth handle medical code translation?

Absolutely. It features tools for mapping and translating between systems like ICD-9, ICD-10, NDC, RxNorm, and ATC.

What is PyHealth used for?

PyHealth is used for building machine learning models for healthcare applications, including predicting clinical outcomes, processing medical codes, and analyzing physiological signals.

What healthcare-specific models are available?

PyHealth includes over 33 models, ranging from general baselines to specialized healthcare architectures like RETAIN for interpretability and SafeDrug for medication safety.

PyHealth AI Toolkit

Name: PyHealth AI Toolkit
Author: Zehong-Wang

byZehong-Wang

0•

데이터 과학 및 ML

Develops and deploys specialized machine learning models for clinical healthcare tasks using electronic health records, medical coding, and physiological data.

PyHealth is a comprehensive healthcare AI skill designed to streamline the development of clinical machine learning models. It provides a modular 5-stage pipeline—covering data loading, task definition, model selection, training, and deployment—specifically optimized for healthcare data. Whether you are working with electronic health records (EHR) like MIMIC-IV, translating between complex medical coding systems such as ICD-10 and ATC, or implementing interpretable deep learning architectures like RETAIN, this skill provides the domain-specific guidance and best practices needed to build reliable, fair, and clinically-validated AI solutions.

주요 기능

01Cross-system medical code translation for ICD, NDC, RxNorm, and ATC coding systems.

0220+ predefined clinical tasks including mortality, readmission, and drug recommendation.

03Integrated clinical evaluation tools for fairness, calibration, and uncertainty quantification.

04Access to 33+ specialized models including RETAIN, SafeDrug, GAMENet, and Transformers for EHR.

05Standardized data loading for major clinical datasets including MIMIC-III/IV, eICU, and OMOP.

060 GitHub stars

사용 사례

01Classifying sleep stages and detecting physiological anomalies from EEG/ECG signals.

02Developing medication recommendation systems that minimize adverse drug-drug interactions.

03Predicting patient mortality and hospital readmission risk using sequential EHR data.

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add zehong-wang/kosmos pyhealth

For use in Claude.ai and ChatGPT

Download Skill