Multi-Center Validation of Video-Based Deep Learning to Evaluate Defecation Patterns on 3D High-Definition Anorectal Manometry.

Background: Deep learning technologies have demonstrated the ability to identify dyssynergic defecation for diagnosis of common gastrointestinal motility disorders through nuanced interpretation of 3-dimensional high definition anal manometry (3D-HDAM). We aimed to validate a deep learning algorithm capable of spatiotemporal analysis of 3D-HDAM in a multi-center setting.

Methods: We included 1,214 consecutive anorectal manometry studies performed across three large healthcare systems between 2018-2022. Deep learning results were compared to expert interpretation according to the London consensus protocol as reference standard. Diagnostic accuracy was assessed using bootstrap sampling to calculate area-under-the-curve (AUC). We used Wilcoxon tests to analyze how well the confidence scores from the deep learning model correlated with the likelihood that experts would assign ambiguous labels in cases where determinations were uncertain. Video-based deep learning features were clustered using Gaussian Mixture Modeling to reveal novel dyssynergia subtypes.

Results: The deep hybrid learning algorithm achieved AUCs of 0.99 (± 0.001 standard deviation), 0.90 ± 0.008, and 0.79 ± 0.003 at Dartmouth Health, Henry Ford Hospital, and University of Pittsburg Medical Center respectively, performance comparable or superior to solely deep learning or traditional modeling on every cohort. The algorithm appeared capable of reporting confidence aligned with manual expert interpretation of ambiguity (W=-20.50 [p<0.001]; -1.73 [p=0.08]; -3.22 [p=0.001]). We further identified two novel classes of dyssynergia patterns that may represent clinically relevant phenotypes of dyssynergia.

Conclusions: 3D high-definition anorectal manometry combined with video-based deep learning is a useful and clinically relevant technology for evaluating anorectal dyssynergia. Future use cases can be expanded to evaluating other motility disorders and their treatment.