Build clinical prediction pipelines with checks that matter.
Builds and checks clinical prediction pipelines with PyHealth, from supervised samples through patient-level splitting, training, and held-out evaluation.
When to use it
Use for clinical prediction with PyHealth involving EHR sequences, physiological signals, imaging, or medical-code lookup.
Give it a clinical prediction task; it builds a PyHealth pipeline with patient partitions, compatible models, validation metrics, and held-out evaluation.
Requires Python 3.12 or 3.13; PyHealth 2.0.2 excludes Python 3.14 and later.
Requires PyHealth 2.0.2, which the smoke-test command installs into an isolated environment.
This skill
arXiv
Checks the paper on arXiv
medical-code tables
Downloads medical-code tables