Efficacy of a comprehensive binary classification model using a deep convolutional neural network for wireless capsule endoscopy

Abstract The manual reading of capsule endoscopy (CE) videos in small bowel disease diagnosis is time-intensive. Algorithms introduced to automate this process are premature for real clinical applications, and multi-diagnosis using these methods has not been sufficiently validated. Therefore, we dev...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Sang Hoon Kim, Youngbae Hwang, Dong Jun Oh, Ji Hyung Nam, Ki Bae Kim, Junseok Park, Hyun Joo Song, Yun Jeong Lim
Formato:	article
Lenguaje:	EN
Publicado:	Nature Portfolio 2021
Materias:	Medicine R Science Q
Acceso en línea:	https://doaj.org/article/c7ec7dda26254e068eb086a14b675a25
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:c7ec7dda26254e068eb086a14b675a25
record_format	dspace
spelling	oai:doaj.org-article:c7ec7dda26254e068eb086a14b675a252021-12-02T19:04:19ZEfficacy of a comprehensive binary classification model using a deep convolutional neural network for wireless capsule endoscopy10.1038/s41598-021-96748-z2045-2322https://doaj.org/article/c7ec7dda26254e068eb086a14b675a252021-09-01T00:00:00Zhttps://doi.org/10.1038/s41598-021-96748-zhttps://doaj.org/toc/2045-2322Abstract The manual reading of capsule endoscopy (CE) videos in small bowel disease diagnosis is time-intensive. Algorithms introduced to automate this process are premature for real clinical applications, and multi-diagnosis using these methods has not been sufficiently validated. Therefore, we developed a practical binary classification model, which selectively identifies clinically meaningful images including inflamed mucosa, atypical vascularity or bleeding, and tested it with unseen cases. Four hundred thousand CE images were randomly selected from 84 cases in which 240,000 images were used to train the algorithm to categorize images binarily. The remaining images were utilized for validation and internal testing. The algorithm was externally tested with 256,591 unseen images. The diagnostic accuracy of the trained model applied to the validation set was 98.067%. In contrast, the accuracy of the model when applied to a dataset provided by an independent hospital that did not participate during training was 85.470%. The area under the curve (AUC) was 0.922. Our model showed excellent internal test results, and the misreadings were slightly increased when the model was tested in unseen external cases while the classified ‘insignificant’ images contain ambiguous substances. Once this limitation is solved, the proposed CNN-based binary classification will be a promising candidate for developing clinically-ready computer-aided reading methods.Sang Hoon KimYoungbae HwangDong Jun OhJi Hyung NamKi Bae KimJunseok ParkHyun Joo SongYun Jeong LimNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 11, Iss 1, Pp 1-11 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Medicine R Science Q
spellingShingle	Medicine R Science Q Sang Hoon Kim Youngbae Hwang Dong Jun Oh Ji Hyung Nam Ki Bae Kim Junseok Park Hyun Joo Song Yun Jeong Lim Efficacy of a comprehensive binary classification model using a deep convolutional neural network for wireless capsule endoscopy
description	Abstract The manual reading of capsule endoscopy (CE) videos in small bowel disease diagnosis is time-intensive. Algorithms introduced to automate this process are premature for real clinical applications, and multi-diagnosis using these methods has not been sufficiently validated. Therefore, we developed a practical binary classification model, which selectively identifies clinically meaningful images including inflamed mucosa, atypical vascularity or bleeding, and tested it with unseen cases. Four hundred thousand CE images were randomly selected from 84 cases in which 240,000 images were used to train the algorithm to categorize images binarily. The remaining images were utilized for validation and internal testing. The algorithm was externally tested with 256,591 unseen images. The diagnostic accuracy of the trained model applied to the validation set was 98.067%. In contrast, the accuracy of the model when applied to a dataset provided by an independent hospital that did not participate during training was 85.470%. The area under the curve (AUC) was 0.922. Our model showed excellent internal test results, and the misreadings were slightly increased when the model was tested in unseen external cases while the classified ‘insignificant’ images contain ambiguous substances. Once this limitation is solved, the proposed CNN-based binary classification will be a promising candidate for developing clinically-ready computer-aided reading methods.
format	article
author	Sang Hoon Kim Youngbae Hwang Dong Jun Oh Ji Hyung Nam Ki Bae Kim Junseok Park Hyun Joo Song Yun Jeong Lim
author_facet	Sang Hoon Kim Youngbae Hwang Dong Jun Oh Ji Hyung Nam Ki Bae Kim Junseok Park Hyun Joo Song Yun Jeong Lim
author_sort	Sang Hoon Kim
title	Efficacy of a comprehensive binary classification model using a deep convolutional neural network for wireless capsule endoscopy
title_short	Efficacy of a comprehensive binary classification model using a deep convolutional neural network for wireless capsule endoscopy
title_full	Efficacy of a comprehensive binary classification model using a deep convolutional neural network for wireless capsule endoscopy
title_fullStr	Efficacy of a comprehensive binary classification model using a deep convolutional neural network for wireless capsule endoscopy
title_full_unstemmed	Efficacy of a comprehensive binary classification model using a deep convolutional neural network for wireless capsule endoscopy
title_sort	efficacy of a comprehensive binary classification model using a deep convolutional neural network for wireless capsule endoscopy
publisher	Nature Portfolio
publishDate	2021
url	https://doaj.org/article/c7ec7dda26254e068eb086a14b675a25
work_keys_str_mv	AT sanghoonkim efficacyofacomprehensivebinaryclassificationmodelusingadeepconvolutionalneuralnetworkforwirelesscapsuleendoscopy AT youngbaehwang efficacyofacomprehensivebinaryclassificationmodelusingadeepconvolutionalneuralnetworkforwirelesscapsuleendoscopy AT dongjunoh efficacyofacomprehensivebinaryclassificationmodelusingadeepconvolutionalneuralnetworkforwirelesscapsuleendoscopy AT jihyungnam efficacyofacomprehensivebinaryclassificationmodelusingadeepconvolutionalneuralnetworkforwirelesscapsuleendoscopy AT kibaekim efficacyofacomprehensivebinaryclassificationmodelusingadeepconvolutionalneuralnetworkforwirelesscapsuleendoscopy AT junseokpark efficacyofacomprehensivebinaryclassificationmodelusingadeepconvolutionalneuralnetworkforwirelesscapsuleendoscopy AT hyunjoosong efficacyofacomprehensivebinaryclassificationmodelusingadeepconvolutionalneuralnetworkforwirelesscapsuleendoscopy AT yunjeonglim efficacyofacomprehensivebinaryclassificationmodelusingadeepconvolutionalneuralnetworkforwirelesscapsuleendoscopy
_version_	1718377236382351360

Efficacy of a comprehensive binary classification model using a deep convolutional neural network for wireless capsule endoscopy

Ejemplares similares