Anorectal disorders are common in general population and often misdiagnosed. These disorders have a significant impact in the quality of life of patients who often are unable to find concrete answers to their complaints. Anorectal manometry is the gold standard for the evaluation of suspected motility disorders. Nonetheless, the interpretation of these exams is hampered by the limited availability of experts in this technique as well as the high degree of variability in procedure protocols. 

Our group, in collaboration with a high-volume center in Brazil (Pelvia – Gastrointestinal Motility and Continence, Curitiba, Paraná, Brazil) developed several machine learning algorithms for the automatic identification and differentiation of anorectal motility patterns, focusing on the distinction between fecal incontinence and obstructed defecation patterns. Overall, the most efficient model achieved an accuracy of 85% and a high discriminating capacity, with an area under the curve of 0.94. Our group was the first to describe the application of an artificial intelligence algorithm to anorectal manometry may pave the way to improve care of patients with anorectal disorders. This study has been accepted for publication on the prestigious journal Clinical and Translational Gastroenterology.

The link to the paper will be made available soon.

In press // Clin Transl Gastroenterol 2022 // December 2022