ID | 63844 |
フルテキストURL | |
著者 |
Sukegawa, Shintaro
Department of Oral Pathology and Medicine, Graduate School of Medicine, Dentistry and Pharmaceutical Sciences, Okayama University
ORCID
Kaken ID
publons
Tanaka, Futa
Department of Electrical, Electronic and Computer Engineering, Faculty of Engineering, Gifu University
Nakano, Keisuke
Department of Oral Pathology and Medicine, Graduate School of Medicine, Dentistry and Pharmaceutical Sciences, Okayama University
ORCID
Kaken ID
publons
researchmap
Hara, Takeshi
Department of Electrical, Electronic and Computer Engineering, Faculty of Engineering, Gifu University
Yoshii, Kazumasa
Department of Electrical, Electronic and Computer Engineering, Faculty of Engineering, Gifu University
Yamashita, Katsusuke
Polytechnic Center Kagawa
Ono, Sawako
Department of Pathology, Kagawa Prefectural Central Hospital
Takabatake, Kiyofumi
Department of Oral Pathology and Medicine, Graduate School of Medicine, Dentistry and Pharmaceutical Sciences, Okayama University
Kaken ID
publons
researchmap
Kawai, Hotaka
Department of Oral Pathology and Medicine, Graduate School of Medicine, Dentistry and Pharmaceutical Sciences, Okayama University
Nagatsuka, Hitoshi
Department of Oral Pathology and Medicine, Graduate School of Medicine, Dentistry and Pharmaceutical Sciences, Okayama University
Kaken ID
publons
researchmap
Furuki, Yoshihiko
Department of Oral and Maxillofacial Surgery, Kagawa Prefectural Central Hospital
|
抄録 | The use of sharpness aware minimization (SAM) as an optimizer that achieves high performance for convolutional neural networks (CNNs) is attracting attention in various fields of deep learning. We used deep learning to perform classification diagnosis in oral exfoliative cytology and to analyze performance, using SAM as an optimization algorithm to improve classification accuracy. The whole image of the oral exfoliation cytology slide was cut into tiles and labeled by an oral pathologist. CNN was VGG16, and stochastic gradient descent (SGD) and SAM were used as optimizers. Each was analyzed with and without a learning rate scheduler in 300 epochs. The performance metrics used were accuracy, precision, recall, specificity, F1 score, AUC, and statistical and effect size. All optimizers performed better with the rate scheduler. In particular, the SAM effect size had high accuracy (11.2) and AUC (11.0). SAM had the best performance of all models with a learning rate scheduler. (AUC = 0.9328) SAM tended to suppress overfitting compared to SGD. In oral exfoliation cytology classification, CNNs using SAM rate scheduler showed the highest classification performance. These results suggest that SAM can play an important role in primary screening of the oral cytological diagnostic environment.
|
発行日 | 2022-08-02
|
出版物タイトル |
Scientific Reports
|
巻 | 12巻
|
号 | 1号
|
出版者 | Nature Portfolio
|
開始ページ | 13281
|
ISSN | 2045-2322
|
資料タイプ |
学術雑誌論文
|
言語 |
英語
|
OAI-PMH Set |
岡山大学
|
著作権者 | © The Author(s) 2022
|
論文のバージョン | publisher
|
PubMed ID | |
DOI | |
Web of Science KeyUT | |
関連URL | isVersionOf https://doi.org/10.1038/s41598-022-17602-4
|
ライセンス | http://creativecommons.org/licenses/by/4.0/
|