pyhealth
Builds and validates PyHealth clinical machine-learning pipelines for EHR
by k-dense-aiFree
Description
pyhealth is an agent skill for building and validating clinical machine-learning pipelines with PyHealth, covering electronic health record sequences, physiological signals, imaging tasks and medical-code lookup. It also covers loading datasets such as MIMIC-III and MIMIC-IV. The pipeline runs Dataset, then Task, then Model, then Trainer, then Metrics. The skill targets PyHealth 2.0.2, verified against the released wheel, the current documentation and CPU runs on Python 3.12 with Torch 2.7.1. Its 1.x and 2.x interfaces differ, and the skill warns not to combine legacy Visit examples with the 2.x event and processor API. Before modelling, the skill asks you to establish the cohort, the prediction time, the observation window, the outcome horizon and the unit of evaluation. For general tabular learning that has no PyHealth dataset or task, the skill is optional. It includes a starter pipeline with a demo mode you can run to check the install. Use it for clinical prediction research with PyHealth. It is part of the k-dense-ai scientific agent skills repository on GitHub, which has about 49,600 stars. The listing shows it as free, and it has about 1,800 installs.…
Highlights
- Dataset, Task, Model, Trainer and Metrics pipeline
- EHR, signal, imaging and medical-code tasks
- MIMIC-III and MIMIC-IV loading
- Starter pipeline with a demo mode
- Targets PyHealth 2.0.2
Ratings & Reviews
No ratings yet
Be the first to rate pyhealth.
People also visit
pyhealth FAQ
What is PyHealth used for?
Clinical machine-learning pipelines for EHR sequences, physiological signals, imaging and medical codes.
Which version is covered?
PyHealth 2.0.2, verified on Python 3.12 with Torch 2.7.1.
Can I mix 1.x and 2.x examples?
No. The skill says not to combine legacy Visit examples with the 2.x event and processor API.
What should be defined before modelling?
The cohort, prediction time, observation window, outcome horizon and unit of evaluation.
How do I install it?
Run npx skills add https://github.com/k-dense-ai/scientific-agent-skills --skill pyhealth.
Is it checked for security?
Its security audits show passes from Gen Agent Trust Hub, Socket and Snyk.
What are the best pyhealth alternatives?
Popular alternatives to pyhealth include healthcare-emr-patterns, clinical-reports and hypogenic. See all pyhealth alternatives