pyhealth

Name: pyhealth
Availability: InStock
Author: HolobiomicsLab

Develop and deploy clinical machine learning models for EHR data.

4|1|Updated Jun 18, 2025

Search Tags:#ai #healthcare #datasets #model-training #interpretability #clinical-ml

Ready to run

npx skills add https://github.com/HolobiomicsLab/Toolomics --skill pyhealth-holobiomicslab

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: pyhealth
Source: https://github.com/HolobiomicsLab/Toolomics/tree/main/mcp_host/skills/scientific-skills/scientific-skills/pyhealth
Command: npx skills add https://github.com/HolobiomicsLab/Toolomics --skill pyhealth-holobiomicslab

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

PyHealth provides a comprehensive, scalable toolkit to develop, test, and deploy machine learning models for healthcare data, reducing the time to translate clinical data into actionable AI solutions.

Core Features & Use Cases

Comprehensive data handling for Electronic Health Records (EHR), clinical tasks (mortality, readmission, length of stay), medical coding (ICD/NDC/ATC), and multi-modal data (signals, imaging, text).
End-to-end pipelines from data loading to model training, evaluation, and interpretability, with modular components for datasets, tasks, models, and trainer.
Real-world use cases include mortality risk prediction, hospital readmission forecasting, drug recommendation with safety constraints, and clinical code translation to harmonize multi-site studies.

Quick Start

Install PyHealth and load a healthcare dataset, define a mortality task, and train a baseline model to start evaluating clinical predictions.

Frequently Asked Questions about pyhealth

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

FAQPage Schema

How do I train machine learning models for clinical prediction tasks using EHR data?▼

To train clinical ML models using EHR data, you can use PyHealth to load structured healthcare datasets, define clinical prediction tasks like mortality or readmission, and execute modular pipelines from preprocessing to model training and evaluation.

What healthcare datasets and data formats does this clinical ML toolkit support?▼

This clinical ML toolkit supports comprehensive data handling for Electronic Health Records (EHR), physiological signals, imaging, and clinical text. It also processes medical coding formats including ICD, NDC, and ATC for multi-modal healthcare datasets.

Can I build drug recommendation models with safety constraints using healthcare AI?▼

Yes, you can build drug recommendation models with safety constraints using this healthcare AI toolkit. PyHealth includes modular components for task-definition and model selection to support real-world drug recommendation forecasting.

Does this toolkit support model interpretability for clinical machine learning?▼

Yes, model interpretability for clinical machine learning is fully supported. PyHealth provides end-to-end pipelines that include evaluation and interpretability components to help translate clinical data into actionable AI solutions.

What is the best way to forecast hospital readmission and length of stay with multi-modal data?▼

The best way to forecast hospital readmission and length of stay is using a modular clinical ML pipeline. PyHealth enables you to load multi-modal data, define the forecasting task, train a baseline model, and evaluate clinical predictions end-to-end.

Do I need structured datasets to deploy clinical machine learning pipelines?▼

Yes, you need structured datasets to deploy clinical machine learning pipelines. PyHealth requires structured healthcare datasets and modular pipeline configurations to support task-definition, preprocessing, model training, and deployment.