Privacy and limits

Privacy: Runs in your browser; files are processed on your device and not uploaded by the app; the page is downloaded from GitHub Pages and the CDN, which see your IP address; synthetic data only; not a validated or Part 11 system.

Limits: Here every detector runs its [SI] stand-in (robust z, 1-D Wasserstein drift, bag-of-words text isolation) with built-in thresholds. On the 150-participant synthetic demo it flagged all 3 injected outliers plus 1 participant with no injected outlier, and exactly the 3 injected unusual notes; with no labeled evaluation, no accuracy was measured. Every flag needs human review. What this does not show.

Starting Python in your browser; first visit downloads about 30 MB.

Loading the page code.

Central statistical monitoring results: flagged per-participant numeric outliers with their extreme columns, a cohort-wide drift table per column, and the start of the per-lab drift table.
Screenshot of this app's results on the bundled synthetic data. To use your own files, run it locally.