Health and clinical
Survival analysis, diagnostic accuracy and agreement on patient-level records, with an AI statistician that runs on your own network and encrypted projects at rest.
Methods
Kaplan–Meier curves, Cox regression and competing risks.
ROC curves and areas under the curve.
Bland-Altman plots and agreement statistics for method comparison.
Logistic, Poisson and other generalized linear models, and mixed models for repeated measurements.
Design-correct estimates from health surveys with weights, strata and clusters.
Normality, equal variances and outliers tested for you, with the right alternative when one fails.
Every module is in every plan, with 148 procedures in all. See how the engine is validated.
Controls
On by default: the AI may reach only model servers on your machine or the hospital network.
AES-256-GCM projects, encrypted autosave and recovery, and password-protected .sav files.
Each AI request records what was shared and what was approved.
IT locks the model endpoint, forbids cloud models and caps what the AI may see, for every install.
A typical project
Open the extract on a hospital machine; nothing is uploaded, and the app works offline for up to 30 days between licence checks.
Point every seat at the hospital's own model server, or run the built-in local model.
Survival, ROC and agreement analyses with their assumption checks.
APA-style tables and figures to Word or PDF for the paper or the trial report.
Every module and the private AI, free for 14 days. Institutions can plan a pilot with us.