利用多重深度学习卷积神经网络提高视频胶囊内窥镜图像分类的准确性

iGIE Pub Date : 2024-03-01 DOI:10.1016/j.igie.2023.11.007
Dongguang Li PhD , David Cave MD, PhD , April Li , Shaoguang Li MD, PhD
{"title":"利用多重深度学习卷积神经网络提高视频胶囊内窥镜图像分类的准确性","authors":"Dongguang Li PhD ,&nbsp;David Cave MD, PhD ,&nbsp;April Li ,&nbsp;Shaoguang Li MD, PhD","doi":"10.1016/j.igie.2023.11.007","DOIUrl":null,"url":null,"abstract":"<div><h3>Background and aims</h3><p>Video capsule endoscopy (VCE) is widely used in the detection of abnormalities in the small intestine. However, it remains challenging to correctly identify a limited number of possible abnormal images from tens of thousands of total images, and this impediment has limited expansion of the technology. More recently, artificial intelligence (AI) technology has been used in classifying VCE images from patients, but clinical-grade diagnostic accuracy (&gt;99%) has not been achieved.</p></div><div><h3>Methods</h3><p>This study proposes a system for the automatic classification of a number of categories of unbounded VCE images with high accuracy by means of a transfer learning approach using multiple convolutional neural networks (CNNs). With this new approach, it is not necessary to implement image segmentation; thus, the feature extraction becomes automatic, and the existing models can be fine-tuned to obtain specific classifiers.</p></div><div><h3>Results</h3><p>More than 16,000 VCE GI images from normal individuals, including those with normal clean mucosa, the pylorus, the ileocecal valve, a reduced mucosal view due to luminal contents and lymphangiectasia (a normal variant), and patients with 5 pathologic states (angioectasia, bleeding, erosions, ulcers, and foreign bodies), were obtained from a publicly available data set. These were used in building, testing, and validating AI models for evaluating the diagnostic accuracy of our combined 17-CNN deep learning approach. Compared with a single CNN approach used by other research groups, our AI method, using 17 CNNs, achieved an overall diagnostic accuracy of 99.79%, with an accuracy of 100% for identifying bleeding and foreign bodies. The high accuracy was further shown in the confusion matrices, precision, recall, and F1 score.</p></div><div><h3>Conclusions</h3><p>We have developed accurate AI deep learning models for unbounded VCE image classification of various medical conditions in medical practice.</p></div>","PeriodicalId":100652,"journal":{"name":"iGIE","volume":"3 1","pages":"Pages 72-81"},"PeriodicalIF":0.0000,"publicationDate":"2024-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S2949708623001371/pdfft?md5=bd53bbbf38460b8793ef89e37ac1f248&pid=1-s2.0-S2949708623001371-main.pdf","citationCount":"0","resultStr":"{\"title\":\"Enhanced accuracy for classification of video capsule endoscopy images using multiple deep learning convolutional neural networks\",\"authors\":\"Dongguang Li PhD ,&nbsp;David Cave MD, PhD ,&nbsp;April Li ,&nbsp;Shaoguang Li MD, PhD\",\"doi\":\"10.1016/j.igie.2023.11.007\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><h3>Background and aims</h3><p>Video capsule endoscopy (VCE) is widely used in the detection of abnormalities in the small intestine. However, it remains challenging to correctly identify a limited number of possible abnormal images from tens of thousands of total images, and this impediment has limited expansion of the technology. More recently, artificial intelligence (AI) technology has been used in classifying VCE images from patients, but clinical-grade diagnostic accuracy (&gt;99%) has not been achieved.</p></div><div><h3>Methods</h3><p>This study proposes a system for the automatic classification of a number of categories of unbounded VCE images with high accuracy by means of a transfer learning approach using multiple convolutional neural networks (CNNs). With this new approach, it is not necessary to implement image segmentation; thus, the feature extraction becomes automatic, and the existing models can be fine-tuned to obtain specific classifiers.</p></div><div><h3>Results</h3><p>More than 16,000 VCE GI images from normal individuals, including those with normal clean mucosa, the pylorus, the ileocecal valve, a reduced mucosal view due to luminal contents and lymphangiectasia (a normal variant), and patients with 5 pathologic states (angioectasia, bleeding, erosions, ulcers, and foreign bodies), were obtained from a publicly available data set. These were used in building, testing, and validating AI models for evaluating the diagnostic accuracy of our combined 17-CNN deep learning approach. Compared with a single CNN approach used by other research groups, our AI method, using 17 CNNs, achieved an overall diagnostic accuracy of 99.79%, with an accuracy of 100% for identifying bleeding and foreign bodies. The high accuracy was further shown in the confusion matrices, precision, recall, and F1 score.</p></div><div><h3>Conclusions</h3><p>We have developed accurate AI deep learning models for unbounded VCE image classification of various medical conditions in medical practice.</p></div>\",\"PeriodicalId\":100652,\"journal\":{\"name\":\"iGIE\",\"volume\":\"3 1\",\"pages\":\"Pages 72-81\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-03-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.sciencedirect.com/science/article/pii/S2949708623001371/pdfft?md5=bd53bbbf38460b8793ef89e37ac1f248&pid=1-s2.0-S2949708623001371-main.pdf\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"iGIE\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S2949708623001371\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"iGIE","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2949708623001371","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

摘要

背景和目的视频胶囊内窥镜(VCE)被广泛应用于小肠异常的检测。然而,要从数以万计的总图像中正确识别数量有限的可能异常图像仍具有挑战性,这一障碍限制了该技术的推广。最近,人工智能(AI)技术已被用于对患者的 VCE 图像进行分类,但临床诊断准确率(99%)尚未达到。方法本研究提出了一种利用多重卷积神经网络(CNN)的迁移学习方法,对无限制的 VCE 图像进行自动分类的系统,准确率很高。采用这种新方法,无需进行图像分割;因此,特征提取变得自动化,而且可以对现有模型进行微调,以获得特定的分类器。结果 从公开数据集中获取了超过 16,000 张正常人的 VCE 胃肠道图像,包括正常清洁粘膜、幽门、回盲瓣、因管腔内容物和淋巴管扩张(正常变异)导致的粘膜视野缩小,以及 5 种病理状态(血管扩张、出血、糜烂、溃疡和异物)的患者。这些数据用于构建、测试和验证人工智能模型,以评估我们的 17-CNN 深度学习组合方法的诊断准确性。与其他研究小组使用的单一 CNN 方法相比,我们的人工智能方法使用了 17 个 CNN,总体诊断准确率达到 99.79%,其中识别出血和异物的准确率为 100%。高准确率进一步体现在混淆矩阵、精确度、召回率和 F1 分数上。结论我们开发了准确的人工智能深度学习模型,用于医疗实践中各种医疗状况的无界 VCE 图像分类。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Enhanced accuracy for classification of video capsule endoscopy images using multiple deep learning convolutional neural networks

Background and aims

Video capsule endoscopy (VCE) is widely used in the detection of abnormalities in the small intestine. However, it remains challenging to correctly identify a limited number of possible abnormal images from tens of thousands of total images, and this impediment has limited expansion of the technology. More recently, artificial intelligence (AI) technology has been used in classifying VCE images from patients, but clinical-grade diagnostic accuracy (>99%) has not been achieved.

Methods

This study proposes a system for the automatic classification of a number of categories of unbounded VCE images with high accuracy by means of a transfer learning approach using multiple convolutional neural networks (CNNs). With this new approach, it is not necessary to implement image segmentation; thus, the feature extraction becomes automatic, and the existing models can be fine-tuned to obtain specific classifiers.

Results

More than 16,000 VCE GI images from normal individuals, including those with normal clean mucosa, the pylorus, the ileocecal valve, a reduced mucosal view due to luminal contents and lymphangiectasia (a normal variant), and patients with 5 pathologic states (angioectasia, bleeding, erosions, ulcers, and foreign bodies), were obtained from a publicly available data set. These were used in building, testing, and validating AI models for evaluating the diagnostic accuracy of our combined 17-CNN deep learning approach. Compared with a single CNN approach used by other research groups, our AI method, using 17 CNNs, achieved an overall diagnostic accuracy of 99.79%, with an accuracy of 100% for identifying bleeding and foreign bodies. The high accuracy was further shown in the confusion matrices, precision, recall, and F1 score.

Conclusions

We have developed accurate AI deep learning models for unbounded VCE image classification of various medical conditions in medical practice.

求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信