Health AI Central
Peer-Reviewed Publication
Brachytherapy2026June 11, 2026Journal Article

Deep learning-based automatic segmentation of rectal tumors in endoscopic images.

Alana Thibodeau-Antonacci1, Aurélie Garant2, Corey Miller3, Té Vuong4, Luca Leon Weishaupt5, Philippe Nicolaï6, Shirin A Enger7
1Medical Physics Unit, Department of Oncology, McGill University, Montréal, Québec, Canada. Electronic address: alana.thibodeau-antonacci@mail.mcgill.ca.
2Department of Radiation Oncology, UT Southwestern Medical Center, Dallas, TX.
3Division of Gastroenterology, Department of Medicine, Jewish General Hospital, Montréal, Québec, Canada; Lady Davis Institute for Medical Research, Jewish General Hospital, Montréal, Québec, Canada.
4Department of Oncology, Jewish General Hospital, Montréal, Québec, Canada.
5Medical Physics Unit, Department of Oncology, McGill University, Montréal, Québec, Canada.
6Centre Lasers Intenses et Applications, CEA CNRS - Université Bordeaux 1, Talence, France.
7Medical Physics Unit, Department of Oncology, McGill University, Montréal, Québec, Canada; Lady Davis Institute for Medical Research, Jewish General Hospital, Montréal, Québec, Canada; Research Institute of McGill University Health Center, Montréal, Québec, Canada.
PubMed DOI

Abstract

PURPOSE: Endoscopy is critical in the identification of rectal tumors, but is prone to observer errors. The aim of this study was to assess the inter- and intra-observer variability in delineating rectal lesions in endoscopic images taken during high-dose-rate (HDR) brachytherapy and develop a deep learning-based automatic tumor segmentation model. MATERIALS AND METHODS: Three expert annotators i…

Create a free account to keep reading

Free members get 10 full research views every month across publications, clinical trials, FDA clearances, adverse events, and NIH grants. No credit card required.

Create Free AccountSign In

Want unlimited research access? See Pro plans

Data Accuracy Notice: Research intelligence on Health AI Central is aggregated from public sources (PubMed, ClinicalTrials.gov, FDA, NIH, CMS, and others) and refreshed nightly. Classifications and derived metrics are produced by automated methods described in our Methodology. We recommend verifying critical data points against the primary sources before making decisions.