Duke researchers develop 'smart toilet' that monitors bowel health

Gastroenterologists typically have to rely on a patient's self-reported information about their stools, which can be unreliable. An artificial intelligence tool being developed by Durham, N.C.-based Duke University researchers will analyze a patient's stool for them.

A May 24 news release said researchers analyzed 3,328 stool images provided by study participants or found online. All of the images were reviewed by gastroenterologists and classified according to the widely used Bristol stool scale. 

Five things to know:

  1. The tool can be fitted into the pipes of an existing toilet. Once a person flushes the toilet, the toilet will take an image of the stool within the pipes. 

  2. Deborah Fisher, MD, associate professor of medicine at Duke and one of the study's lead authors, said that normally, gastroenterologists rely on patients' self-reported information on their stool to inform medical treatment, but this can be unreliable. Often, patients can't remember what their stool looked like or how often they had a bowel movement, which is a vital part of monitoring progress. 

  3. It can give gastroenterologists information to inform their treatment for chronic issues, such as inflammatory bowel disease or irritable bowel syndrome. It can also improve the timeliness of diagnosing these conditions, and the tool can monitor a patient's response to medications.

  4. Researchers found the AI tool accurately classified stool type 85 percent of the time and blood detection was accurate 76 percent of the time.

  5. The AI tool is not available to the public yet. Researchers are adding features to the technology so that stool specimen sampling for biochemical marker analysis provides specific disease data.

Copyright © 2024 Becker's Healthcare. All Rights Reserved. Privacy Policy. Cookie Policy. Linking and Reprinting Policy.

 

Featured Whitepapers

Featured Webinars