Real-Time Multi-Task Deep Learning Model for Polyp Detection, Characterization, and Size Estimation
While performing a colonoscopy, there are many tasks to be done: finding polyps, classifying them, and deciding the next procedure for the polyps, whether to incise them or not. Such tasks are challenging for fellow doctors. All these three tasks can have an intrapersonal error, which varies among e...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10835090/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832592836245061632 |
|---|---|
| author | Phanukorn Sunthornwetchapong Kasichon Hombubpha Kasenee Tiankanon Satimai Aniwan Pasit Jakkrawankul Natawut Nupairoj Peerapon Vateekul Rungsun Rerknimitr |
| author_facet | Phanukorn Sunthornwetchapong Kasichon Hombubpha Kasenee Tiankanon Satimai Aniwan Pasit Jakkrawankul Natawut Nupairoj Peerapon Vateekul Rungsun Rerknimitr |
| author_sort | Phanukorn Sunthornwetchapong |
| collection | DOAJ |
| description | While performing a colonoscopy, there are many tasks to be done: finding polyps, classifying them, and deciding the next procedure for the polyps, whether to incise them or not. Such tasks are challenging for fellow doctors. All these three tasks can have an intrapersonal error, which varies among endoscopists. A proven method for enhancing performance is computer-aided detection and a diagnosis system for endoscopists, which tends to be a real-time system. In this work, we present a modified convolutional neural network (CNN) based deep learning (DL) model to perform these tasks in real-time, utilizing existing object detection models: YOLOv5 and YOLOv8. For the various tasks, the models are trained using datasets with incomplete labels, leading to a comparison of different training strategies. Our model, YOLOv8, achieved an F1-score of 95.96% for the polyp detection task, 85.24% F1-score for the polyp classification task, and 78.41% macro F1-score for the polyp size estimation task. Such results, when compared with fellow doctors’ findings proved superior in both accuracy and macro F1-score, maintaining a real-time inference speed. |
| format | Article |
| id | doaj-art-1094ac8b7c29443ebc60d6290f6e0e51 |
| institution | Kabale University |
| issn | 2169-3536 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-1094ac8b7c29443ebc60d6290f6e0e512025-01-21T00:02:17ZengIEEEIEEE Access2169-35362025-01-01138469848110.1109/ACCESS.2025.352772010835090Real-Time Multi-Task Deep Learning Model for Polyp Detection, Characterization, and Size EstimationPhanukorn Sunthornwetchapong0https://orcid.org/0000-0002-6142-8268Kasichon Hombubpha1Kasenee Tiankanon2Satimai Aniwan3Pasit Jakkrawankul4Natawut Nupairoj5Peerapon Vateekul6https://orcid.org/0000-0001-9718-3592Rungsun Rerknimitr7https://orcid.org/0000-0001-6866-6886Department of Computer Engineering, Faculty of Engineering, Chulalongkorn University, Pathum Wan, Bangkok, ThailandDepartment of Medicine, Faculty of Medicine, King Chulalongkorn Memorial Hospital, Division of Gastroenterology, Chulalongkorn University, Bangkok, ThailandDepartment of Medicine, Faculty of Medicine, King Chulalongkorn Memorial Hospital, Division of Gastroenterology, Chulalongkorn University, Bangkok, ThailandDepartment of Medicine, Faculty of Medicine, King Chulalongkorn Memorial Hospital, Division of Gastroenterology, Chulalongkorn University, Bangkok, ThailandDepartment of Computer Engineering, Faculty of Engineering, Chulalongkorn University, Pathum Wan, Bangkok, ThailandDepartment of Computer Engineering, Faculty of Engineering, Chulalongkorn University, Pathum Wan, Bangkok, ThailandDepartment of Computer Engineering, Faculty of Engineering, Chulalongkorn University, Pathum Wan, Bangkok, ThailandDepartment of Medicine, Faculty of Medicine, King Chulalongkorn Memorial Hospital, Division of Gastroenterology, Chulalongkorn University, Bangkok, ThailandWhile performing a colonoscopy, there are many tasks to be done: finding polyps, classifying them, and deciding the next procedure for the polyps, whether to incise them or not. Such tasks are challenging for fellow doctors. All these three tasks can have an intrapersonal error, which varies among endoscopists. A proven method for enhancing performance is computer-aided detection and a diagnosis system for endoscopists, which tends to be a real-time system. In this work, we present a modified convolutional neural network (CNN) based deep learning (DL) model to perform these tasks in real-time, utilizing existing object detection models: YOLOv5 and YOLOv8. For the various tasks, the models are trained using datasets with incomplete labels, leading to a comparison of different training strategies. Our model, YOLOv8, achieved an F1-score of 95.96% for the polyp detection task, 85.24% F1-score for the polyp classification task, and 78.41% macro F1-score for the polyp size estimation task. Such results, when compared with fellow doctors’ findings proved superior in both accuracy and macro F1-score, maintaining a real-time inference speed.https://ieeexplore.ieee.org/document/10835090/Colonic polypdeep learningreal-time image classificationreal-time object detectionreal-time size estimation |
| spellingShingle | Phanukorn Sunthornwetchapong Kasichon Hombubpha Kasenee Tiankanon Satimai Aniwan Pasit Jakkrawankul Natawut Nupairoj Peerapon Vateekul Rungsun Rerknimitr Real-Time Multi-Task Deep Learning Model for Polyp Detection, Characterization, and Size Estimation IEEE Access Colonic polyp deep learning real-time image classification real-time object detection real-time size estimation |
| title | Real-Time Multi-Task Deep Learning Model for Polyp Detection, Characterization, and Size Estimation |
| title_full | Real-Time Multi-Task Deep Learning Model for Polyp Detection, Characterization, and Size Estimation |
| title_fullStr | Real-Time Multi-Task Deep Learning Model for Polyp Detection, Characterization, and Size Estimation |
| title_full_unstemmed | Real-Time Multi-Task Deep Learning Model for Polyp Detection, Characterization, and Size Estimation |
| title_short | Real-Time Multi-Task Deep Learning Model for Polyp Detection, Characterization, and Size Estimation |
| title_sort | real time multi task deep learning model for polyp detection characterization and size estimation |
| topic | Colonic polyp deep learning real-time image classification real-time object detection real-time size estimation |
| url | https://ieeexplore.ieee.org/document/10835090/ |
| work_keys_str_mv | AT phanukornsunthornwetchapong realtimemultitaskdeeplearningmodelforpolypdetectioncharacterizationandsizeestimation AT kasichonhombubpha realtimemultitaskdeeplearningmodelforpolypdetectioncharacterizationandsizeestimation AT kaseneetiankanon realtimemultitaskdeeplearningmodelforpolypdetectioncharacterizationandsizeestimation AT satimaianiwan realtimemultitaskdeeplearningmodelforpolypdetectioncharacterizationandsizeestimation AT pasitjakkrawankul realtimemultitaskdeeplearningmodelforpolypdetectioncharacterizationandsizeestimation AT natawutnupairoj realtimemultitaskdeeplearningmodelforpolypdetectioncharacterizationandsizeestimation AT peeraponvateekul realtimemultitaskdeeplearningmodelforpolypdetectioncharacterizationandsizeestimation AT rungsunrerknimitr realtimemultitaskdeeplearningmodelforpolypdetectioncharacterizationandsizeestimation |