Which healthcare datasets does PyHealth support?

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

Can this skill help with medical code translation?

Yes, it provides specialized mapping tools (CrossMap and InnerMap) for translating between systems like ICD-9/10, NDC, RxNorm, and ATC.

Does it include specialized healthcare neural network architectures?

Yes, it implements 33+ models including healthcare-specific architectures like RETAIN for interpretability, SafeDrug for medication safety, and StageNet for EHR sequences.

Is PyHealth suitable for predicting patient outcomes?

Absolutely. It includes predefined functions for critical tasks such as mortality prediction, length of stay estimation, and 30-day readmission risk.

PyHealth: Healthcare AI Toolkit

Name: PyHealth: Healthcare AI Toolkit
Author: Activer007

byActiver007

Ciencia de Datos y ML

Simplifies the development and deployment of machine learning models for clinical data and healthcare applications.

Acerca de

PyHealth is a comprehensive Python library designed for healthcare AI, offering specialized tools for processing electronic health records (EHR), medical coding systems, and physiological signals. It provides a modular five-stage pipeline for data loading, task definition, and training state-of-the-art models like RETAIN and SafeDrug, enabling researchers and developers to build robust clinical prediction models for tasks such as mortality prediction and drug recommendation while ensuring model interpretability, fairness, and clinical validity.

Características Principales

Implementation of 33+ deep learning models specialized for sequential clinical data
Comprehensive medical coding translation across ICD, NDC, RxNorm, and ATC systems
Integrated evaluation tools for model calibration, fairness assessment, and interpretability
Standardized interfaces for 10+ healthcare datasets including MIMIC-III/IV, eICU, and OMOP
0 GitHub stars
Predefined clinical prediction tasks for mortality, readmission, and length of stay

Casos de Uso

Building medication recommendation systems with drug-drug interaction constraints
Developing predictive models for hospital readmission and ICU mortality risk
Standardizing heterogeneous medical codes across different healthcare databases

Acerca de

Características Principales

Implementation of 33+ deep learning models specialized for sequential clinical data
Comprehensive medical coding translation across ICD, NDC, RxNorm, and ATC systems
Integrated evaluation tools for model calibration, fairness assessment, and interpretability
Standardized interfaces for 10+ healthcare datasets including MIMIC-III/IV, eICU, and OMOP
0 GitHub stars
Predefined clinical prediction tasks for mortality, readmission, and length of stay

Casos de Uso

Building medication recommendation systems with drug-drug interaction constraints
Developing predictive models for hospital readmission and ICU mortality risk
Standardizing heterogeneous medical codes across different healthcare databases