Abstract
dc:descriptionPeyronie’s disease (PD) is a chronic fibrotic condition characterized by plaque formation in the tunica albuginea, leading to penile deformity. Beyond its physical impact, PD causes significant psychological distress and relationship difficulties, and it remains heavily underreported, with only a fraction of affected patients seeking clinical evaluation or treatment. Current assessment methods typically rely on invasive procedures involving medication-induced erections or on 2D photography, both of which have notable limitations. In particular, 2D images fail to capture the full three-dimensional nature of penile deformities, reducing their effectiveness for comprehensive PD assessment. To address these gaps, this thesis explores the use of artificial intelligence (AI)-based 3D reconstruction methods to develop an automated pipeline for PD assessment. Leveraging recent advances in generative models and 3D computer vision, we aim to reconstruct accurate 3D representations of the penile shaft from multiple 2D images, enabling a more comprehensive analysis of curvature and deformity. As part of this work, we create 246 Computer-Aided Design (CAD) models simulating various PD presentations, encompassing a wide range of curvature angles and complex multiplanar deformities. A subset of these models is fabricated using 3D printers to further evaluate the pipeline in a realistic setting. Systematic protocols for photographing and rendering these models are established to ensure robust inputs for the 3D reconstruction. We employ TRELLIS, a state-of-the-art multi-view 3D generative model, to reconstruct 3D penile shapes from the captured images. These reconstructions are used in our automated curvature analysis tool, PenoMeter, to assess PD by analyzing renders of the 3D models. This pipeline not only advances the technical capability to model PD in three dimensions but also lays the groundwork for developing clinically viable tools. Ultimately, our work demonstrates that integrating AI-based 3D reconstruction into PD evaluation holds significant promise for delivering objective, reliable, and accessible assessments that can be seamlessly incorporated into clinical workflows, improving both diagnosis and patient care.
Degree
thesis:*- Name thesis:degree_name
- Master of Science - MSc
- Level thesis:degree_level
- master's
- Discipline thesis:degree_discipline
- Bioinformatics
- Grantor dc:publisher
- University of British Columbia
- Year dc:date
- 2025
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Balapour, Ali
- Contributors dc:contributor
-
- Hach, Faraz
- Flannigan, Ryan
Rights
dc:rights- Statement dc:rights
-
- Attribution-NonCommercial-NoDerivatives 4.0 International
- Language dc:language
- eng
Identifiers
dc:identifier.*- Handle dc:identifier
- http://hdl.handle.net/2429/92161