Cargando…

Effectiveness of deep learning classifiers in histopathological diagnosis of oral squamous cell carcinoma by pathologists

The study aims to identify histological classifiers from histopathological images of oral squamous cell carcinoma using convolutional neural network (CNN) deep learning models and shows how the results can improve diagnosis. Histopathological samples of oral squamous cell carcinoma were prepared by...

Descripción completa

Detalles Bibliográficos
Autores principales: Sukegawa, Shintaro, Ono, Sawako, Tanaka, Futa, Inoue, Yuta, Hara, Takeshi, Yoshii, Kazumasa, Nakano, Keisuke, Takabatake, Kiyofumi, Kawai, Hotaka, Katsumitsu, Shimada, Nakai, Fumi, Nakai, Yasuhiro, Miyazaki, Ryo, Murakami, Satoshi, Nagatsuka, Hitoshi, Miyake, Minoru
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Nature Publishing Group UK 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10356919/
https://www.ncbi.nlm.nih.gov/pubmed/37468501
http://dx.doi.org/10.1038/s41598-023-38343-y
_version_ 1785075381346238464
author Sukegawa, Shintaro
Ono, Sawako
Tanaka, Futa
Inoue, Yuta
Hara, Takeshi
Yoshii, Kazumasa
Nakano, Keisuke
Takabatake, Kiyofumi
Kawai, Hotaka
Katsumitsu, Shimada
Nakai, Fumi
Nakai, Yasuhiro
Miyazaki, Ryo
Murakami, Satoshi
Nagatsuka, Hitoshi
Miyake, Minoru
author_facet Sukegawa, Shintaro
Ono, Sawako
Tanaka, Futa
Inoue, Yuta
Hara, Takeshi
Yoshii, Kazumasa
Nakano, Keisuke
Takabatake, Kiyofumi
Kawai, Hotaka
Katsumitsu, Shimada
Nakai, Fumi
Nakai, Yasuhiro
Miyazaki, Ryo
Murakami, Satoshi
Nagatsuka, Hitoshi
Miyake, Minoru
author_sort Sukegawa, Shintaro
collection PubMed
description The study aims to identify histological classifiers from histopathological images of oral squamous cell carcinoma using convolutional neural network (CNN) deep learning models and shows how the results can improve diagnosis. Histopathological samples of oral squamous cell carcinoma were prepared by oral pathologists. Images were divided into tiles on a virtual slide, and labels (squamous cell carcinoma, normal, and others) were applied. VGG16 and ResNet50 with the optimizers stochastic gradient descent with momentum and spectral angle mapper (SAM) were used, with and without a learning rate scheduler. The conditions for achieving good CNN performances were identified by examining performance metrics. We used ROCAUC to statistically evaluate diagnostic performance improvement of six oral pathologists using the results from the selected CNN model for assisted diagnosis. VGG16 with SAM showed the best performance, with accuracy = 0.8622 and AUC = 0.9602. The diagnostic performances of the oral pathologists statistically significantly improved when the diagnostic results of the deep learning model were used as supplementary diagnoses (p-value = 0.031). By considering the learning results of deep learning model classifiers, the diagnostic accuracy of pathologists can be improved. This study contributes to the application of highly reliable deep learning models for oral pathological diagnosis.
format Online
Article
Text
id pubmed-10356919
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher Nature Publishing Group UK
record_format MEDLINE/PubMed
spelling pubmed-103569192023-07-21 Effectiveness of deep learning classifiers in histopathological diagnosis of oral squamous cell carcinoma by pathologists Sukegawa, Shintaro Ono, Sawako Tanaka, Futa Inoue, Yuta Hara, Takeshi Yoshii, Kazumasa Nakano, Keisuke Takabatake, Kiyofumi Kawai, Hotaka Katsumitsu, Shimada Nakai, Fumi Nakai, Yasuhiro Miyazaki, Ryo Murakami, Satoshi Nagatsuka, Hitoshi Miyake, Minoru Sci Rep Article The study aims to identify histological classifiers from histopathological images of oral squamous cell carcinoma using convolutional neural network (CNN) deep learning models and shows how the results can improve diagnosis. Histopathological samples of oral squamous cell carcinoma were prepared by oral pathologists. Images were divided into tiles on a virtual slide, and labels (squamous cell carcinoma, normal, and others) were applied. VGG16 and ResNet50 with the optimizers stochastic gradient descent with momentum and spectral angle mapper (SAM) were used, with and without a learning rate scheduler. The conditions for achieving good CNN performances were identified by examining performance metrics. We used ROCAUC to statistically evaluate diagnostic performance improvement of six oral pathologists using the results from the selected CNN model for assisted diagnosis. VGG16 with SAM showed the best performance, with accuracy = 0.8622 and AUC = 0.9602. The diagnostic performances of the oral pathologists statistically significantly improved when the diagnostic results of the deep learning model were used as supplementary diagnoses (p-value = 0.031). By considering the learning results of deep learning model classifiers, the diagnostic accuracy of pathologists can be improved. This study contributes to the application of highly reliable deep learning models for oral pathological diagnosis. Nature Publishing Group UK 2023-07-19 /pmc/articles/PMC10356919/ /pubmed/37468501 http://dx.doi.org/10.1038/s41598-023-38343-y Text en © The Author(s) 2023 https://creativecommons.org/licenses/by/4.0/Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ (https://creativecommons.org/licenses/by/4.0/) .
spellingShingle Article
Sukegawa, Shintaro
Ono, Sawako
Tanaka, Futa
Inoue, Yuta
Hara, Takeshi
Yoshii, Kazumasa
Nakano, Keisuke
Takabatake, Kiyofumi
Kawai, Hotaka
Katsumitsu, Shimada
Nakai, Fumi
Nakai, Yasuhiro
Miyazaki, Ryo
Murakami, Satoshi
Nagatsuka, Hitoshi
Miyake, Minoru
Effectiveness of deep learning classifiers in histopathological diagnosis of oral squamous cell carcinoma by pathologists
title Effectiveness of deep learning classifiers in histopathological diagnosis of oral squamous cell carcinoma by pathologists
title_full Effectiveness of deep learning classifiers in histopathological diagnosis of oral squamous cell carcinoma by pathologists
title_fullStr Effectiveness of deep learning classifiers in histopathological diagnosis of oral squamous cell carcinoma by pathologists
title_full_unstemmed Effectiveness of deep learning classifiers in histopathological diagnosis of oral squamous cell carcinoma by pathologists
title_short Effectiveness of deep learning classifiers in histopathological diagnosis of oral squamous cell carcinoma by pathologists
title_sort effectiveness of deep learning classifiers in histopathological diagnosis of oral squamous cell carcinoma by pathologists
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10356919/
https://www.ncbi.nlm.nih.gov/pubmed/37468501
http://dx.doi.org/10.1038/s41598-023-38343-y
work_keys_str_mv AT sukegawashintaro effectivenessofdeeplearningclassifiersinhistopathologicaldiagnosisoforalsquamouscellcarcinomabypathologists
AT onosawako effectivenessofdeeplearningclassifiersinhistopathologicaldiagnosisoforalsquamouscellcarcinomabypathologists
AT tanakafuta effectivenessofdeeplearningclassifiersinhistopathologicaldiagnosisoforalsquamouscellcarcinomabypathologists
AT inoueyuta effectivenessofdeeplearningclassifiersinhistopathologicaldiagnosisoforalsquamouscellcarcinomabypathologists
AT haratakeshi effectivenessofdeeplearningclassifiersinhistopathologicaldiagnosisoforalsquamouscellcarcinomabypathologists
AT yoshiikazumasa effectivenessofdeeplearningclassifiersinhistopathologicaldiagnosisoforalsquamouscellcarcinomabypathologists
AT nakanokeisuke effectivenessofdeeplearningclassifiersinhistopathologicaldiagnosisoforalsquamouscellcarcinomabypathologists
AT takabatakekiyofumi effectivenessofdeeplearningclassifiersinhistopathologicaldiagnosisoforalsquamouscellcarcinomabypathologists
AT kawaihotaka effectivenessofdeeplearningclassifiersinhistopathologicaldiagnosisoforalsquamouscellcarcinomabypathologists
AT katsumitsushimada effectivenessofdeeplearningclassifiersinhistopathologicaldiagnosisoforalsquamouscellcarcinomabypathologists
AT nakaifumi effectivenessofdeeplearningclassifiersinhistopathologicaldiagnosisoforalsquamouscellcarcinomabypathologists
AT nakaiyasuhiro effectivenessofdeeplearningclassifiersinhistopathologicaldiagnosisoforalsquamouscellcarcinomabypathologists
AT miyazakiryo effectivenessofdeeplearningclassifiersinhistopathologicaldiagnosisoforalsquamouscellcarcinomabypathologists
AT murakamisatoshi effectivenessofdeeplearningclassifiersinhistopathologicaldiagnosisoforalsquamouscellcarcinomabypathologists
AT nagatsukahitoshi effectivenessofdeeplearningclassifiersinhistopathologicaldiagnosisoforalsquamouscellcarcinomabypathologists
AT miyakeminoru effectivenessofdeeplearningclassifiersinhistopathologicaldiagnosisoforalsquamouscellcarcinomabypathologists