Deep learning model applied to real-time delineation of colorectal polyps

Abstract Background Deep learning models have shown considerable potential to improve diagnostic accuracy across medical fields. Although YOLACT has demonstrated real-time detection and segmentation in non-medical datasets, its application in medical settings remains underexplored. This study evalua...

Full description

Saved in:
Bibliographic Details
Main Authors: Moana Gelu-Simeon, Adel Mamou, Georgette Saint-Georges, Marceline Alexis, Marie Sautereau, Yassine Mamou, Jimmy Simeon
Format: Article
Language:English
Published: BMC 2025-06-01
Series:BMC Medical Informatics and Decision Making
Subjects:
Online Access:https://doi.org/10.1186/s12911-025-03047-y
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849724876956368896
author Moana Gelu-Simeon
Adel Mamou
Georgette Saint-Georges
Marceline Alexis
Marie Sautereau
Yassine Mamou
Jimmy Simeon
author_facet Moana Gelu-Simeon
Adel Mamou
Georgette Saint-Georges
Marceline Alexis
Marie Sautereau
Yassine Mamou
Jimmy Simeon
author_sort Moana Gelu-Simeon
collection DOAJ
description Abstract Background Deep learning models have shown considerable potential to improve diagnostic accuracy across medical fields. Although YOLACT has demonstrated real-time detection and segmentation in non-medical datasets, its application in medical settings remains underexplored. This study evaluated the performance of a YOLACT-derived Real-time Polyp Delineation Model (RTPoDeMo) for real-time use on prospectively recorded colonoscopy videos. Methods: Twelve combinations of architectures, including Mask-RCNN, YOLACT, and YOLACT++, paired with backbones such as ResNet50, ResNet101, and DarkNet53, were tested on 2,188 colonoscopy images with three image resolution sizes. Dataset preparation involved pre-processing and segmentation annotation, with optimized image augmentation. Results: RTPoDeMo, using YOLACT-ResNet50, achieved 72.3 mAP and 32.8 FPS for real-time instance segmentation based on COCO annotations. The model performed with a per-image accuracy of 99.59% (95% CI: [99.45 − 99.71%]), sensitivity of 90.63% (95% CI: [78.95 − 93.64%]), specificity of 99.95% (95% CI: [99.93 − 99.97%]) and a F1-score of 0.94 (95% CI: [0.87–0.98]). In validation, out of 36 polyps detected by experts, RTPoDeMo missed only one polyp, compared to six missed by senior endoscopists. The model demonstrated good agreement with experts, reflected by a Cohen’s Kappa coefficient of 0.72 (95% CI: [0.54–1.00], p < 0.0001). Conclusions: Our model provides new perspectives in the adaptation of YOLACT to the real-time delineation of colorectal polyps. In the future, it could improve the characterization of polyps to be resected during colonoscopy.
format Article
id doaj-art-ab92cb09caf3456cac6501bf62d4e3ca
institution DOAJ
issn 1472-6947
language English
publishDate 2025-06-01
publisher BMC
record_format Article
series BMC Medical Informatics and Decision Making
spelling doaj-art-ab92cb09caf3456cac6501bf62d4e3ca2025-08-20T03:10:37ZengBMCBMC Medical Informatics and Decision Making1472-69472025-06-0125111410.1186/s12911-025-03047-yDeep learning model applied to real-time delineation of colorectal polypsMoana Gelu-Simeon0Adel Mamou1Georgette Saint-Georges2Marceline Alexis3Marie Sautereau4Yassine Mamou5Jimmy Simeon6Service d’Hépato-Gastroentérologie, CHU de la GuadeloupeBiostatistic department, Univ. MontpellierService d’Hépato-Gastroentérologie, CHU de la GuadeloupeService d’Hépato-Gastroentérologie, CHU de la GuadeloupeService d’Hépato-Gastroentérologie, CHU de la GuadeloupeService de Médecine Nucléaire, CHU de la GuadeloupeInformatic Department, CHU de la GuadeloupeAbstract Background Deep learning models have shown considerable potential to improve diagnostic accuracy across medical fields. Although YOLACT has demonstrated real-time detection and segmentation in non-medical datasets, its application in medical settings remains underexplored. This study evaluated the performance of a YOLACT-derived Real-time Polyp Delineation Model (RTPoDeMo) for real-time use on prospectively recorded colonoscopy videos. Methods: Twelve combinations of architectures, including Mask-RCNN, YOLACT, and YOLACT++, paired with backbones such as ResNet50, ResNet101, and DarkNet53, were tested on 2,188 colonoscopy images with three image resolution sizes. Dataset preparation involved pre-processing and segmentation annotation, with optimized image augmentation. Results: RTPoDeMo, using YOLACT-ResNet50, achieved 72.3 mAP and 32.8 FPS for real-time instance segmentation based on COCO annotations. The model performed with a per-image accuracy of 99.59% (95% CI: [99.45 − 99.71%]), sensitivity of 90.63% (95% CI: [78.95 − 93.64%]), specificity of 99.95% (95% CI: [99.93 − 99.97%]) and a F1-score of 0.94 (95% CI: [0.87–0.98]). In validation, out of 36 polyps detected by experts, RTPoDeMo missed only one polyp, compared to six missed by senior endoscopists. The model demonstrated good agreement with experts, reflected by a Cohen’s Kappa coefficient of 0.72 (95% CI: [0.54–1.00], p < 0.0001). Conclusions: Our model provides new perspectives in the adaptation of YOLACT to the real-time delineation of colorectal polyps. In the future, it could improve the characterization of polyps to be resected during colonoscopy.https://doi.org/10.1186/s12911-025-03047-yArtificial intelligenceConvolutional neural networkDeep learning modelInstance segmentation
spellingShingle Moana Gelu-Simeon
Adel Mamou
Georgette Saint-Georges
Marceline Alexis
Marie Sautereau
Yassine Mamou
Jimmy Simeon
Deep learning model applied to real-time delineation of colorectal polyps
BMC Medical Informatics and Decision Making
Artificial intelligence
Convolutional neural network
Deep learning model
Instance segmentation
title Deep learning model applied to real-time delineation of colorectal polyps
title_full Deep learning model applied to real-time delineation of colorectal polyps
title_fullStr Deep learning model applied to real-time delineation of colorectal polyps
title_full_unstemmed Deep learning model applied to real-time delineation of colorectal polyps
title_short Deep learning model applied to real-time delineation of colorectal polyps
title_sort deep learning model applied to real time delineation of colorectal polyps
topic Artificial intelligence
Convolutional neural network
Deep learning model
Instance segmentation
url https://doi.org/10.1186/s12911-025-03047-y
work_keys_str_mv AT moanagelusimeon deeplearningmodelappliedtorealtimedelineationofcolorectalpolyps
AT adelmamou deeplearningmodelappliedtorealtimedelineationofcolorectalpolyps
AT georgettesaintgeorges deeplearningmodelappliedtorealtimedelineationofcolorectalpolyps
AT marcelinealexis deeplearningmodelappliedtorealtimedelineationofcolorectalpolyps
AT mariesautereau deeplearningmodelappliedtorealtimedelineationofcolorectalpolyps
AT yassinemamou deeplearningmodelappliedtorealtimedelineationofcolorectalpolyps
AT jimmysimeon deeplearningmodelappliedtorealtimedelineationofcolorectalpolyps