AI Smartphone App Assists in Assessing Stool Form

A recent study published in the American Journal of Gastroenterology highlights a smartphone app developed by researchers and clinicians at the Cedars–Sinai Medical Center. This app uses AI to assist in stool form assessment and has proven to be more successful than traditional patient self-reporting.  

This study included patients between 18 and 75 years old who had irritable bowel syndrome with diarrhea (IBS-D) and were participating in a new drug therapy clinical trial.  

According to the American College of Gastroenterology, irritable bowel syndrome (IBS) is a functional problem in the bowel that may cause symptoms such as bloating, gas, and so on. In the United States, IBS affects between 10 and 15% of the population, with women being more affected than men.  

Many professionals value stool form as a method of assessing the severity of IBS and use it to develop treatment plans. The researchers in this study state that “because IBS is defined by either diarrhea (in the case of IBS-D) or constipation (in the case of IBS-C), assessments of stool form and frequency are critical to determine the success or benefits of interventions in these conditions.”  

The most common method of assessing stool form is using the Bristol Stool Scale (BSS). According to the Continence Foundation of Australia, the BSS is a seven-part scale used to categorize patient stools. Most commonly, this scale is used by patients to self-report data to physicians. 

Exclusionary criteria were used from the original pool of potential participants to narrow the sample size. Eligible patients were offered the use of the smartphone app to monitor their symptoms, and 39 were accepted.  

The AI in the smartphone app captured, isolated, and analyzed pictures of the patient’s stool. The app was “trained” using manual inputs by professionals to evaluate stool consistency, fragmentation, edge fuzziness, volume, and BSS score.  

Throughout the two week period in which patients were monitored, participants analyzed their stool form manually and reported their BSS. Simultaneously, they used the app to capture, record, and analyze stool form and the frequency of bowel movements.  

The results of this study were able to conclude that the AI could similarly assess the stool form like gastroenterologists can. Additionally, the AI was more sensitive and specific in BSS scoring. AI averaged 16% and 11% higher in sensitivity and specificity, respectively, and had a 6% higher accuracy in diagnosis.  

Furthermore, the publication discussed subject satisfaction with the application finding that, of the 20 participants who responded, “50% described their experience as ‘very pleasant and easy’ and the remaining 50% as ‘somewhat easy and pleasant.’” 

“Additional stool features such as stool consistency, edge fuzziness, fragmentation, and volume may provide added objective benefits in studying GI disorders. This app may become a new and valuable tool for use in trials of luminal GI diseases involving changes in stool form,” concluded the researchers in the publication.