Engineers have developed an artificial intelligence capable of detecting from the noises coming from the toilets if the person is suffering from diarrhea. The goal? Spot the signs of epidemics such as cholera.
- We speak of diarrhea when the stools are liquid or soft, larger and more frequent than usual (more than 3 stools per day).
- Cholera is a serious digestive infection caused by ingesting food or water contaminated with the bacillus Vibrio cholerae. It causes diarrhea and severe dehydration.
- According to Public Health France, there are between 1.4 to 4.3 million cases of cholera worldwide each year. The disease is responsible for 28,000 to 142,000 deaths.
the “diarrhea detector” was presented at the Acoustical Society of America conference held in Nashville from December 5-9, 2022. Alexis Noel, a biomechanical engineering researcher at the Georgia Tech Research Institute, came up with the idea of developing this intelligence artificial after hearing that Covid-19 could be detected in sewage.
“I was curious if we could detect diarrhea using sound, as some people are a little wary of having a camera pointed at their butt in the toilet”explained the co-responsible for the project, taken up by the site webmd.
350 defecation sound samples analyzed
To develop this technology, the researchers collected 350 audio samples of toilet sounds available on YouTube and Soundsnap. Some clips had up to 10 hours of diarrhea sound. Scientists had to listen to them to establish their veracity. A significant experience for the team. “We didn’t know these people, we didn’t know how they were recording, so we had to listen to quite a bit of it. There were definitely a lot of fart noises where we were like, ‘That’s not a fart, that’s is someone who blows in his elbow”explains Maia Gatlin, co-leader of the project, an aerospace engineer and doctoral student at the Georgia Tech Research Institute.
But this rather thankless work paid off. The prototype accurately identified diarrhea 98% of the time when tested. Even with background noise, the efficiency was 96%.
Diarrhea detector: how does it work?
The device takes a 10 second audio recording. “We don’t collect anything identifiable about people”, assures Maia Gatlin. The sounds of defecation, urination, flatulence and diarrhea are converted by the device into spectrograms (diagram representing the noise spectrum over time) and then transmitted to the algorithm. Based on the data collected throughout its development and use, the artificial intelligence estimates whether the person is suffering from diarrhea or not. This process only takes a few seconds.
American scientists have developed this technology to help identify the beginnings of cholera epidemics, a deadly disease causing major digestive disorders. “Cholera usually has a more watery sound – it can sound a lot like urination and it doesn’t have a lot of flatulence sounds in general”explains Maia Gatlin.
Researchers have yet to determine how many of these devices would be needed to cover a community, or where the ideal placement would be to detect cases. However, public restrooms could be a possibility.
Artificial intelligence could also eventually be integrated into smart home devices to monitor gut health. Nevertheless, the team specifies that the algorithm and the device still need to be improved (remote report sending, tests with noises obtained under controlled conditions, sounds coming from outside latrines, etc.).
