A team of Delhi's Indraprastha Institute of Information Technology has developed an apparatus that can predict bowel movements to help people with lower body paralysis make arrangements in advance.
The apparatus consisting of a stethoscope and a microphone has to be wrapped around the abdomen. The stethoscope will listen to the intestinal sounds while the microphone will amplify it and send it to a smartphone, which will record these sounds.
"Using a pre-trained machine learning classifier, the sounds will be classified as either a bowel sound or a non-bowel sound in the smartphone. In case if the sound is a bowel sound, the person will be notified that s/he is about to experience a bowel movement," assistant professor at the institute's Computer Science department and team member Juhi Ranjan said.
"Due to the muscle function loss, people with lower body paralysis find it difficult to sense their bowel timings which may sometime result in bowel accidents making it a situation of embarrassment for them. To avoid such awkward situations, they tend to follow a very restricted lifestyle which includes strictly monitoring the food they should eat, the timings of food intake, using an enema, etc."
The project is a creative blend of internet of things and machine learning techniques and aims to help such patients live a healthy and less-restricted life in terms of the food they take, she added.
Other student members of the team are Anmolpreet Kaur, Yashika Arora, Kritika Bansal, and Kunal Suryavanshi.
The project is right now in testing phase and the team aims to test it on at least 100 users before it is made available for commercial use.
"Right now accuracy rate is 95 per cent and after completion of testing phase, we are expecting 100 percent accuracy. Possibly by the end of this year, the product will be available for commercial use," Ranjan said.
With availability of smartphones, the apparatus may cost somewhere around Rs 1,000, they said.
(This story has not been edited by Business Standard staff and is auto-generated from a syndicated feed.)