Software

Explore PyHealth, our comprehensive Python library for healthcare predictive modeling and machine learning applications.

PyHealth

Healthcare ML Library

PyHealth

A Comprehensive Python Library for Health Predictive Models

Python

MIT License

PyHealth is a comprehensive Python library designed for developing and evaluating health predictive models using electronic health records (EHR). It provides a unified framework for healthcare machine learning research and applications, featuring implementations of various algorithms for clinical prediction, medical image processing, and healthcare data analysis.

Key Features:

• Comprehensive healthcare data processing pipelines
• State-of-the-art predictive models for clinical applications
• Support for various healthcare data formats (FHIR, OMOP, etc.)
• Interpretable machine learning models for clinical decision support
• Extensive documentation and tutorials
• Active community and regular updates

Applications:

• Clinical risk prediction and early warning systems
• Drug discovery and molecular property prediction
• Healthcare resource optimization
• Population health management
• Clinical trial patient matching

Documentation GitHub PyPI

Quick Start

Get started with PyHealth in just a few steps

Installation

pip install pyhealth

Basic Usage

from pyhealth.datasets import MIMIC3Dataset
from pyhealth.tasks import drug_recommendation_mimic3_fn
from pyhealth.models import Transformer

# Load dataset
dataset = MIMIC3Dataset(root="/path/to/mimic3")
dataset = dataset.set_task(drug_recommendation_mimic3_fn)

# Initialize model
model = Transformer(dataset=dataset)

# Train and evaluate
model.fit(train_dataset)
model.evaluate(test_dataset)

Citation

If you use PyHealth in your research, please cite our work

@article{zhao2021pyhealth,
  title={PyHealth: A Python Library for Health Predictive Models},
  author={Zhao, Yue and Qiao, Zhi and Xiao, Cao and Glass, Lucas and Sun, Jimeng},
  journal={arXiv preprint arXiv:2101.04209},
  year={2021}
}