UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

A machine learning and algorithmic tool to advance peyronie’s disease assessment Soltani, Reza

Abstract

Peyronie's disease (PD) is a debilitating condition where the development of penile fibrosis leads to dysfunctional deformities, curvature, shortening, and painful erections. The prevalence rate of PD ranges from 0.3% to 20.3%, yet only 0.5-1.3% of adult males present to a healthcare provider. PD often impairs their ability to be sexually active and leads to depression and relationship challenges. Standard of care evaluation for those who seek medical attention includes performing an office-based penile curvature assessment that utilizes a goniometer and ruler to evaluate the curve, length, and other deformities after a stimulated erection by injecting a medical agent. This procedure is cumbersome, painful, and a barrier to appropriate treatment decisions. Alternatively, at-home penile morphological assessment is an approach to address these shortcomings. Patients can send a photograph of their erected penis to a healthcare provider for evaluation and treatment planning. Here, we developed an AI powered assistive diagnostic toolkit that can automatically assess the curvature angle and location given the penile 2D images to be utilized by health care practitioners and the general public. Its performance in finding the point of maximal curvature and its angle is on par with three sub-specialized physicians. Furthermore, its performance is objective and can be utilized for tracking the improvement of PD in time-series data.

Item Media

Item Citations and Data

Rights

Attribution-NonCommercial-NoDerivatives 4.0 International