What is the PyHealth Claude Code Skill?

It is a specialized capability for Claude Code that provides domain-specific guidance and implementation patterns for the PyHealth library, focusing on healthcare-specific AI development.

Does it support medical code standardization?

Yes, it includes comprehensive tools for mapping and translating between medical coding systems such as ICD-9/10, NDC, RxNorm, and ATC.

What specialized clinical models can I implement?

You can implement over 33 models, including healthcare-specific architectures like RETAIN for interpretability, SafeDrug for safe prescriptions, and StageNet for longitudinal data.

How does it handle model evaluation in a clinical context?

It provides specialized tools for evaluating models beyond simple accuracy, including clinical fairness metrics, calibration methods, and uncertainty quantification.

Which healthcare datasets are supported?

The skill supports over 10 major datasets, including MIMIC-III, MIMIC-IV, eICU, OMOP, and various physiological signal datasets like SleepEDF and SHHS.

PyHealth Healthcare AI

Name: PyHealth Healthcare AI
Author: ricable

byricable

•

Data Science & ML

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

About

PyHealth is a comprehensive healthcare AI skill designed to streamline the development of clinical machine learning models through a modular five-stage pipeline. It provides Claude with domain-specific expertise for processing electronic health records (EHR), translating complex medical coding systems like ICD and NDC, and implementing specialized deep learning architectures such as RETAIN and SafeDrug. Whether you are working with large-scale datasets like MIMIC-IV or processing physiological signals like EEG, this skill ensures best practices in clinical data handling, model calibration, and fairness evaluation for healthcare applications.

Key Features

Automated medical coding translation across ICD-9/10, NDC, RxNorm, and ATC systems.
Modular pipeline for data loading, task definition, model training, and clinical evaluation.
Standardized processing for 10+ major healthcare datasets including MIMIC, eICU, and OMOP.
Advanced clinical metrics for fairness, calibration, and uncertainty quantification.
1 GitHub stars
Access to 33+ pre-built clinical models including healthcare-specific Transformers and GNNs.

Use Cases

Processing physiological signals for automated medical tasks like sleep staging or seizure detection from EEG data.
Designing medication recommendation systems that integrate drug-drug interaction (DDI) constraints for patient safety.
Predicting clinical outcomes such as mortality, hospital readmission, and length of stay from longitudinal EHR data.

About

Key Features

Automated medical coding translation across ICD-9/10, NDC, RxNorm, and ATC systems.
Modular pipeline for data loading, task definition, model training, and clinical evaluation.
Standardized processing for 10+ major healthcare datasets including MIMIC, eICU, and OMOP.
Advanced clinical metrics for fairness, calibration, and uncertainty quantification.
1 GitHub stars
Access to 33+ pre-built clinical models including healthcare-specific Transformers and GNNs.

Use Cases

Processing physiological signals for automated medical tasks like sleep staging or seizure detection from EEG data.
Designing medication recommendation systems that integrate drug-drug interaction (DDI) constraints for patient safety.
Predicting clinical outcomes such as mortality, hospital readmission, and length of stay from longitudinal EHR data.