The instrument is intended to be a surrogate for patients’ subjective stool form and frequency using the seven-point Bristol Stool Scale (BSS), which grades consistency from hard to liquid but can produce highly variable results.
It is well known that the subjectivity of BSS can make it difficult to obtain an accurate picture of therapeutic interventions designed to improve symptoms in gastrointestinal disorders such as IBS in clinical trials.
In the study of 39 subjects mentioned In the American Journal of Gastroenterology, researchers at Cedars-Sinai Medical Center in the US asked patients with diarrhea-predominant IBS to take a picture of each stool over a two-week period using the app.
These images were evaluated by AI for five visual characteristics – BSS, consistency, segmentation, edge opacity and size – with the results checked twice by twastroenterologists. Another component of the trial compared the AO with the patient-reported BSS scores.
Overall, there was ‘good agreement’ between AI and expert results across all five measures, and the application outperformed self-reports using the BSS system, with 16% and 11% higher sensitivity and specificity, respectively, resulting in fewer false reports – False positive and negative results.
The authors note that studying gastrointestinal disorders can be challenging, particularly those considered ‘functional’ and rely on patients reporting symptoms such as abdominal pain, bloating, and stool shape and frequency.
There has been an evolution in the measures used in clinical trials of new treatments for IBS, from weekly assessment of symptoms to daily self-report using BSS.
However, BSS “although intuitive, requires objective instruction and guidelines to avoid confusion,” they wrote in the paper.