Build clinical machine learning pipelines from raw patient data to evaluated predictions.
npx skills add https://github.com/dralkh/seerai --skill pyhealth-dralkh
Please help me install this Agent Skill. Skill: pyhealth Source: https://github.com/dralkh/seerai/tree/main/skills/pyhealth Command: npx skills add https://github.com/dralkh/seerai --skill pyhealth-dralkh
PyHealth removes the friction from building healthcare machine learning pipelines by turning raw clinical data into supervised samples, trainable models, and measurable results through one consistent workflow.
Ask for a PyHealth pipeline for your dataset and prediction task, and I will map the right dataset, task, model, split strategy, and evaluation metric for you.
High-intent search queries and answers about installing and using this skill.
To build a clinical ML pipeline, you map your raw MIMIC or eICU data through a Dataset → Task → Model → Trainer → Metrics workflow to generate supervised samples and evaluated predictions. This enforces patient-level splitting and dataset-specific task matching for reproducible research.
You can run healthcare prediction tasks such as mortality prediction, readmission, length of stay, and drug recommendation using electronic health records. The pipeline matches dataset-specific tasks with the correct monitor metric for binary, multiclass, or multilabel learning.
Yes, you can use the same unified workflow for physiological signals and chest X-ray imaging. It applies the consistent Dataset → Task → Model → Trainer → Metrics pattern across EHRs, SleepEDF, EEG, and imaging data to produce measurable prediction results.
You normalize and cross-map medical codes like ICD, ATC, NDC, RxNorm, and CCS using built-in medical code utilities. This process cleans labels and creates more interpretable outputs for ICD coding and code normalization tasks.
The correct way to split patient data for clinical machine learning models is to use patient-level splitting. This prevents data leakage across training and test sets, ensuring that the evaluated predictions from the Trainer and Metrics stages remain statistically valid.
Yes, PyHealth supports sleep staging and EEG event prediction workflows using datasets like SleepEDF. It maps the physiological signal data through the standard pipeline to train models and evaluate predictions with the appropriate multiclass monitor metrics.
jasrajtulsiBuild clinical deep-learning pipelines for healthcare datasets in PyHealth.
williamzujkowskiMitigate security vulnerabilities in ML model deployment workflows.
Estrella-231Build clinical deep-learning pipelines with Dataset, Task, Model, Trainer, and Metrics.
shushuznBuild clinical deep learning pipelines using PyHealth’s unified workflow.
