skills.rest
rubensliv
rubenslivCOMMUNITY
Explore all skills by this author →
GitHub

pyhealth

Build, train, and deploy clinical prediction models on EHR, imaging, and text data.

15|2|Updated Dec 17, 2025
Search Tags:#ai#healthcare#ehr#datasets#model-training#clinical-prediction
One-click install
npx skills add https://github.com/rubensliv/k-dense-ai --skill pyhealth-rubensliv
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: pyhealth
Source: https://github.com/rubensliv/k-dense-ai/tree/main/scientific-skills/pyhealth
Command: npx skills add https://github.com/rubensliv/k-dense-ai --skill pyhealth-rubensliv

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

PyHealth provides a comprehensive healthcare AI toolkit to build, train, and deploy clinical prediction models on real-world EHR, imaging, and text data.

Core Features & Use Cases

  • Datasets and task orchestration: access MIMIC-III/IV, eICU, OMOP, Sleep EEG, and clinical notes; set up sample datasets with task functions to standardize inputs and outputs.
  • End-to-end workflow: preprocessing, model selection (RETAIN, Transformer, GNN), training, evaluation, calibration, and interpretability, enabling rapid prototyping and production-ready deployment.
  • Task-driven collaboration: predefined clinical prediction tasks (mortality, readmission, LOS, drug recommendation) and support for custom task creation to align with clinical questions.

Quick Start

Set up a dataset (e.g., MIMIC-IV), apply the mortality_prediction_mimic4_fn task, train a Transformer model, and evaluate on a held-out test set.

Frequently Asked Questions about pyhealth

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

FAQPage Schema
How do I build clinical prediction models using EHR datasets like MIMIC-IV?▼

To build clinical prediction models using EHR datasets like MIMIC-IV, you apply predefined task functions such as mortality prediction to standardize inputs, then use the toolkit's modular preprocessing, training, and evaluation pipelines to deploy the model.

What healthcare AI tasks can I train on using real-world clinical data?▼

You can train healthcare AI tasks including mortality prediction, readmission, length of stay estimation, and drug recommendation using real-world EHR, imaging, and text data through standardized task schemas and predefined clinical functions.

Which model architectures are available for training clinical prediction models?▼

Available model architectures for training clinical prediction models include Transformer, RETAIN, and GNN, which integrate directly into the end-to-end workflow encompassing preprocessing, training, evaluation, and interpretability.

Can I use clinical notes and Sleep EEG data for model training alongside EHR?▼

Yes, you can use clinical notes and Sleep EEG data for model training alongside EHR datasets like MIMIC-III and eICU, supported by standardized task definitions and modular preprocessing pipelines for multimodal healthcare AI workflows.

Does this healthcare AI toolkit support custom clinical prediction task creation?▼

Yes, the healthcare AI toolkit supports custom clinical prediction task creation, allowing you to define specific task functions that align with your clinical questions while maintaining standardized inputs and outputs for production-grade deployments.

Related Skills

Explore More Data & Analytics Skills→
yf8578yf8578
1

pyhealth

Develop and evaluate healthcare machine learning models using clinical datasets.

Community
Advanced
HolobiomicsLabHolobiomicsLab
4

pyhealth

Develop and deploy clinical machine learning models for EHR data.

Official
Advanced
JosephWoodallJosephWoodall
1

pyhealth

Automate healthcare AI pipelines for EHR data with PyTorch models.

Community
Advanced
silversteinsilverstein
21

pyhealth

Build and evaluate healthcare prediction models using standardized clinical datasets.

Community
Advanced