EHR Risk Research workflow
End-to-end loop so you can finish a methods-style study in the workbench: health → train → trust pack → leakage → external validation → curves / ZIP / predict session. For research and education only. Outputs are not clinical recommendations and are not intended for patient care. We are working toward broader general-purpose use in the future.
Free demo server — it may be slow. Check it with a small amount of data. For larger workloads or freer experimentation, run locally or on your own server. Open live demo
Recommended path
- Open the workbench → Research (or Home → Start research wizard).
- Pick demo/upload CSV + task → run health.
- Train (optional bootstrap samples for ROC/PR CIs on Train).
- Confirm trust pack; run leakage and optional external validation.
- Export ZIP / methods.md; open Analytics for ROC/PR/calibration PNG or Print after retrain.
- Optional: Predict → Export session JSON for lab notebooks.
Key artifacts
| Artifact | Where |
|---|---|
| Trust pack | Per-run file reports/runs/<id>/trust_pack.json (also on run detail; not /v1/reports/trust-pack) |
| Curves + CIs | evaluation_report.json · GET /v1/reports/curves · Analytics |
| Methods + ZIP | Results / wizard download (run_id) |
| Analysis pack | Analytics KPIs · GET /v1/reports/analysis-pack |