Establishing an AI model and application for automated capsule endoscopy recognition based on convolutional neural networks (with video)

Abstract Background Although capsule endoscopy (CE) is a crucial tool for diagnosing small bowel diseases, the need to process a vast number of images imposes a significant workload on physicians, leading to a high risk of missed diagnoses. This study aims to develop an artificial intelligence (AI)...

Full description

Saved in:
Bibliographic Details
Main Authors: Jian Chen, Kaijian Xia, Zihao Zhang, Yu Ding, Ganhong Wang, Xiaodan Xu
Format: Article
Language:English
Published: BMC 2024-11-01
Series:BMC Gastroenterology
Subjects:
Online Access:https://doi.org/10.1186/s12876-024-03482-7
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846171907020292096
author Jian Chen
Kaijian Xia
Zihao Zhang
Yu Ding
Ganhong Wang
Xiaodan Xu
author_facet Jian Chen
Kaijian Xia
Zihao Zhang
Yu Ding
Ganhong Wang
Xiaodan Xu
author_sort Jian Chen
collection DOAJ
description Abstract Background Although capsule endoscopy (CE) is a crucial tool for diagnosing small bowel diseases, the need to process a vast number of images imposes a significant workload on physicians, leading to a high risk of missed diagnoses. This study aims to develop an artificial intelligence (AI) model and application based on convolutional neural networks that can automatically recognize various lesions in small bowel capsule endoscopy. Methods Three small bowel capsule endoscopy datasets were used for AI model training, validation, and testing, encompassing 12 categories of images. The model’s performance was evaluated using metrics such as AUC, sensitivity, specificity, precision, accuracy, and F1 score to select the best model. A human-machine comparison experiment was conducted using the best model and endoscopists with varying levels of experience. Model interpretability was analyzed using Grad-CAM and SHAP techniques. Finally, a clinical application was developed based on the best model using PyQt5 technology. Results A total of 34,303 images were included in this study. The best model, MobileNetv3-large, achieved a weighted average sensitivity of 87.17%, specificity of 98.77%, and an AUC of 0.9897 across all categories. The application developed based on this model performed exceptionally well in comparison with endoscopists, achieving an accuracy of 87.17% and a processing speed of 75.04 frames per second, surpassing endoscopists of varying experience levels. Conclusion The AI model and application developed based on convolutional neural networks can quickly and accurately identify 12 types of small bowel lesions. With its high sensitivity, this system can effectively assist physicians in interpreting small bowel capsule endoscopy images.Future studies will validate the AI system for video evaluations and real-world clinical integration.
format Article
id doaj-art-ae90953c1da74458b673f85c50e6b2da
institution Kabale University
issn 1471-230X
language English
publishDate 2024-11-01
publisher BMC
record_format Article
series BMC Gastroenterology
spelling doaj-art-ae90953c1da74458b673f85c50e6b2da2024-11-10T12:28:11ZengBMCBMC Gastroenterology1471-230X2024-11-0124111710.1186/s12876-024-03482-7Establishing an AI model and application for automated capsule endoscopy recognition based on convolutional neural networks (with video)Jian Chen0Kaijian Xia1Zihao Zhang2Yu Ding3Ganhong Wang4Xiaodan Xu5Department of Gastroenterology, Changshu Hospital Affiliated to Soochow UniversityCenter of Intelligent Medical Technology Research, Changshu Hospital Affiliated to Soochow UniversityShanghai Haoxiong Education Technology Co., Ltd.Department of Gastroenterology, Changshu Hospital Affiliated to Soochow UniversityDepartment of Gastroenterology, Changshu Hospital Affiliated to Nanjing University of Chinese MedicineDepartment of Gastroenterology, Changshu Hospital Affiliated to Soochow UniversityAbstract Background Although capsule endoscopy (CE) is a crucial tool for diagnosing small bowel diseases, the need to process a vast number of images imposes a significant workload on physicians, leading to a high risk of missed diagnoses. This study aims to develop an artificial intelligence (AI) model and application based on convolutional neural networks that can automatically recognize various lesions in small bowel capsule endoscopy. Methods Three small bowel capsule endoscopy datasets were used for AI model training, validation, and testing, encompassing 12 categories of images. The model’s performance was evaluated using metrics such as AUC, sensitivity, specificity, precision, accuracy, and F1 score to select the best model. A human-machine comparison experiment was conducted using the best model and endoscopists with varying levels of experience. Model interpretability was analyzed using Grad-CAM and SHAP techniques. Finally, a clinical application was developed based on the best model using PyQt5 technology. Results A total of 34,303 images were included in this study. The best model, MobileNetv3-large, achieved a weighted average sensitivity of 87.17%, specificity of 98.77%, and an AUC of 0.9897 across all categories. The application developed based on this model performed exceptionally well in comparison with endoscopists, achieving an accuracy of 87.17% and a processing speed of 75.04 frames per second, surpassing endoscopists of varying experience levels. Conclusion The AI model and application developed based on convolutional neural networks can quickly and accurately identify 12 types of small bowel lesions. With its high sensitivity, this system can effectively assist physicians in interpreting small bowel capsule endoscopy images.Future studies will validate the AI system for video evaluations and real-world clinical integration.https://doi.org/10.1186/s12876-024-03482-7Capsule endoscopyArtificial intelligenceApplicationConvolutional neural networksPyQt5
spellingShingle Jian Chen
Kaijian Xia
Zihao Zhang
Yu Ding
Ganhong Wang
Xiaodan Xu
Establishing an AI model and application for automated capsule endoscopy recognition based on convolutional neural networks (with video)
BMC Gastroenterology
Capsule endoscopy
Artificial intelligence
Application
Convolutional neural networks
PyQt5
title Establishing an AI model and application for automated capsule endoscopy recognition based on convolutional neural networks (with video)
title_full Establishing an AI model and application for automated capsule endoscopy recognition based on convolutional neural networks (with video)
title_fullStr Establishing an AI model and application for automated capsule endoscopy recognition based on convolutional neural networks (with video)
title_full_unstemmed Establishing an AI model and application for automated capsule endoscopy recognition based on convolutional neural networks (with video)
title_short Establishing an AI model and application for automated capsule endoscopy recognition based on convolutional neural networks (with video)
title_sort establishing an ai model and application for automated capsule endoscopy recognition based on convolutional neural networks with video
topic Capsule endoscopy
Artificial intelligence
Application
Convolutional neural networks
PyQt5
url https://doi.org/10.1186/s12876-024-03482-7
work_keys_str_mv AT jianchen establishinganaimodelandapplicationforautomatedcapsuleendoscopyrecognitionbasedonconvolutionalneuralnetworkswithvideo
AT kaijianxia establishinganaimodelandapplicationforautomatedcapsuleendoscopyrecognitionbasedonconvolutionalneuralnetworkswithvideo
AT zihaozhang establishinganaimodelandapplicationforautomatedcapsuleendoscopyrecognitionbasedonconvolutionalneuralnetworkswithvideo
AT yuding establishinganaimodelandapplicationforautomatedcapsuleendoscopyrecognitionbasedonconvolutionalneuralnetworkswithvideo
AT ganhongwang establishinganaimodelandapplicationforautomatedcapsuleendoscopyrecognitionbasedonconvolutionalneuralnetworkswithvideo
AT xiaodanxu establishinganaimodelandapplicationforautomatedcapsuleendoscopyrecognitionbasedonconvolutionalneuralnetworkswithvideo