De Carvalho, Thomas;
(2025)
New automated segment classification, withdrawal time measurement, phase detection, and explainable diagnosis methods for colonoscopy using deep learning.
Doctoral thesis (Ph.D), UCL (University College London).
Preview |
Text
De Carvalho_Thesis.pdf - Submitted Version Download (42MB) | Preview |
Abstract
Colonoscopy is widely regarded as the gold standard for the early diagnosis of colorectal cancer. Numerous methods for detecting and diagnosing polyps have been developed, with many achieving important success. Computer-aided detection and diagnosis systems are increasingly being integrated into real-time procedures. Traditionally, the quality of a colonoscopy has been primarily assessed by the Adenoma Detection Rate (ADR), but other important metrics also contribute to evaluating a procedure’s quality. These include withdrawal time, the number of resections, cecal intubation, and retroflexion. However, these metrics are often manually recorded by the clinician, which can be time-consuming and sometimes overlooked. Utilizing these metrics as quality indicators offers several benefits, such as identifying clinicians who may need additional training, providing a consistent method for comparing procedure quality, and generating more data for future research. To realise these benefits, automatic and consistent measurement methods are essential. This thesis addresses this challenge by offering solutions for several key metrics: cecum detection, withdrawal time measurement, colonic segment classification, detection of cleaning and therapeutic phases, retroflexion detection, and classification of out-ofbody segments. Additionally, a novel approach to polyp diagnosis using medically explainable features, rather than a simple final diagnosis, is proposed to enhance clinician training and improve the explainability of computer-aided diagnosis systems. All these methods are designed to be automated and easily integrated into existing polyp detection systems and medical devices
| Type: | Thesis (Doctoral) |
|---|---|
| Qualification: | Ph.D |
| Title: | New automated segment classification, withdrawal time measurement, phase detection, and explainable diagnosis methods for colonoscopy using deep learning |
| Open access status: | An open access version is available from UCL Discovery |
| Language: | English |
| Additional information: | Copyright © The Author 2025. Original content in this thesis is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) Licence (https://creativecommons.org/licenses/by-nc/4.0/). Any third-party copyright material present remains the property of its respective owner(s) and is licensed under its existing terms. Access may initially be restricted at the author’s request. |
| UCL classification: | UCL > Provost and Vice Provost Offices > UCL BEAMS UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science > Dept of Med Phys and Biomedical Eng UCL |
| URI: | https://discovery.ucl.ac.uk/id/eprint/10210436 |
Archive Staff Only
 |
View Item |
