What Is a Stool Scan? How It Works and What It Shows

A stool scan is a photo of your stool analyzed by software: computer vision reads the visual features in the frame — consistency, color, size, visible details — and turns them into structured, dated readings. No kit, no lab, no mailing anything; the whole thing happens on your phone in seconds.

The term gets used loosely, so here’s the precise version: what a stool scan actually reads, what the research says about scanning versus eyeballing, and the honest line between what a photo can show and what only a lab can measure.

What is a stool scan?

A stool scan applies the same family of technology that recognizes faces in your camera roll to a single photo of the toilet bowl. The software classifies where the stool falls on the Bristol stool scale — the seven-type consistency scale developed and validated in 1997 as a practical stand-in for intestinal transit time — places its color on a range, and notes visible details like fragmentation, size, and undigested food. Each scan becomes a stamped entry in a running record, which is where the real value lives: one reading is trivia, a few weeks of readings is a baseline. Our full explainer walks the pipeline step by step.

What does a stool scan show?

• Consistency, typed. The Bristol classification is the headline reading, and it carries real information: in a study of 110 adults, stool form correlated strongly with measured whole-gut transit time (r = −0.61) — while stool frequency correlated poorly. What it looks like says more than how often it happens.
• Color, placed on a range. From pale through very dark, so the occasional detour from your usual stands out against a recorded baseline instead of a guess.
• The details eyes skip. Size and volume, fragmentation, edge definition, visible mucus or undigested food — the things nobody consistently notes on their own.
• Derived estimates. Because form tracks transit, a scan can estimate a transit range and feed simple indicators tied to everyday inputs like hydration and fiber.

Is a stool scan more accurate than just looking?

This has actually been tested, and the answer is yes — with a twist. In a 2022 study in the American Journal of Gastroenterology, researchers had people photograph every stool for two weeks, then compared AI readings of those photos against two expert gastroenterologists. Agreement between the AI and the experts was good to excellent across consistency, fragmentation, and volume (correlation coefficients 0.78–0.89). Agreement between the AI and the participants’ own self-reports was only moderate — and when readings disagreed, the AI matched symptoms better than the people did.

Human self-typing isn’t hopeless — validation studies put overall accuracy around 81% — but the errors cluster exactly at the borderline types where descriptions start mattering. Software applies the same criteria every time, which is the whole argument. We go deeper in how accurate are poop scanner apps.

What can’t a stool scan show?

Anything invisible. Occult blood is literally defined by the fact that you can’t see it just by looking; the same goes for inflammation markers, infections, and enzyme levels — those live in lab stool analysis, ordered through a clinician, and no photo replaces them. A scan reads the visible layer, brilliantly and repeatably, and that’s all it claims. A scan is also not a diagnosis: it’s structured observation, and health concerns belong with a healthcare professional.

Trying it

If you want the manual, no-download version first, our free Poop Check tool walks consistency and color by hand. The automatic version is Stoolio (here’s why we built it): point your phone at the bowl, and the scan types the Bristol form, places the color, reads the visible details, estimates a transit range, and rolls everything into one Stoolio Score — a stool diary that keeps itself, privately.