Build clinical deep learning pipelines using PyHealth’s unified workflow.
npx skills add https://github.com/shushuzn/Rairos --skill pyhealth-shushuzn
Please help me install this Agent Skill. Skill: pyhealth Source: https://github.com/shushuzn/Rairos/tree/main/skills/pyhealth Command: npx skills add https://github.com/shushuzn/Rairos --skill pyhealth-shushuzn
This skill solves the pain of building end-to-end clinical machine learning pipelines from EHR/signal/imaging data without manually wiring datasets, tasks, models, training loops, and metrics.
Use the pyhealth skill to help you assemble a full pipeline for predicting mortality on MIMIC-III by choosing the right dataset, task, model, Trainer monitor metric, and evaluation metrics.
High-intent search queries and answers about installing and using this skill.
Building a clinical ML pipeline for EHR data requires assembling a unified Dataset, Task, Model, Trainer, and Metrics workflow. This approach ensures consistent processing for clinical prediction tasks like mortality and readmission.
To predict patient mortality using MIMIC-III, you need to select a compatible dataset and task pair, instantiate a model with the resulting SampleDataset, apply patient-safe splitting, and train using a task-appropriate clinical metric.
Cross-mapping medical codes like ICD, ATC, NDC, and RxNorm uses built-in medical code utilities. These tools look up and translate code systems for clinical data labeling and feature engineering tasks.
Patient-level data splitting is required for clinical deep learning to prevent data leakage. You must use patient-safe splitting techniques when generating your SampleDataset before training models.
A single unified workflow supports sleep staging and drug recommendation tasks. You can process physiological signals or EHR data by selecting compatible dataset and task pairs within the same pipeline structure.
A unified clinical ML workflow avoids manually wiring datasets, tasks, models, and metrics. It provides practical guidance on critical pitfalls like selecting the correct Dataset type for models and computing task-appropriate clinical metrics.
ovachieverConvert raw clinical data and task specifications into model-ready healthcare AI inputs.
amdCreate export and evaluation plans for Quark quantized models.
Zeyuyang-0420Automate 5-stage clinical deep learning pipelines with PyHealth and PyTorch.
Yezez9Develop, test, and deploy healthcare machine learning models with clinical data.
