The data & inference layer for gut-health products
PoopCheck powers AI stool analysis for consumer apps, wellness programs, and research teams. Integrate our API, or license the labeled dataset it was built on.
Stool Analysis API
A REST API that turns a stool photo into a structured health report in seconds — the same engine that powers our live demo.
- Bristol Stool Scale classification (Types 1–7)
- Color, consistency, and shape analysis
- Wellness indicators for mucus and visible blood
Annotated Dataset
Train your own models, validate research, or benchmark against the largest consumer stool image dataset we know of.
- 185,000+ labeled stool images
- AI labels: Bristol type, colour, blood and mucus
- Custom subsets and formats on request
How our labels are made
AI labels at scale. People check the work.
Three AI models label each photo. Our team then reviews photos, removes the unusable ones and corrects labels, and those corrections go back into training.
-
PoopCheck app
People photograph their stool at home and add daily context.
-
AI labeling
Three models label Bristol type, color, blood, mucus and more, each with a confidence score.
-
Human review Non-medical
Our team removes unusable photos and corrects the labels the models got wrong.
-
Training data
Reviewed photos and corrected labels are what our in-house model learns from.
Corrections retrain our model
Why teams build on PoopCheck
We've been shipping stool-analysis AI to real users since 2024. Every photo goes through three AI models that vote on the result, and our database holds real-world stool photos taken at every angle, in every light and on every kind of phone. That's the engine you'd be plugging into, or the data you'd be learning from.