Multi-Task Deep Learning Model for Classification of Dental Implant Brand and Treatment Stage Using Dental Panoramic Radiograph Images

It is necessary to accurately identify dental implant brands and the stage of treatment to ensure efficient care. Thus, the purpose of this study was to use multi-task deep learning to investigate a classifier that categorizes implant brands and treatment stages from dental panoramic radiographic im...

Full description

Bibliographic Details
Main Authors: Shintaro Sukegawa, Kazumasa Yoshii, Takeshi Hara, Tamamo Matsuyama, Katsusuke Yamashita, Keisuke Nakano, Kiyofumi Takabatake, Hotaka Kawai, Hitoshi Nagatsuka, Yoshihiko Furuki
Format: Article
Language:English
Published: MDPI AG 2021-05-01
Series:Biomolecules
Subjects:
Online Access:https://www.mdpi.com/2218-273X/11/6/815
id doaj-b53532cc20b04af3a072d8d38c1bd5f9
record_format Article
spelling doaj-b53532cc20b04af3a072d8d38c1bd5f92021-06-01T01:41:39ZengMDPI AGBiomolecules2218-273X2021-05-011181581510.3390/biom11060815Multi-Task Deep Learning Model for Classification of Dental Implant Brand and Treatment Stage Using Dental Panoramic Radiograph ImagesShintaro Sukegawa0Kazumasa Yoshii1Takeshi Hara2Tamamo Matsuyama3Katsusuke Yamashita4Keisuke Nakano5Kiyofumi Takabatake6Hotaka Kawai7Hitoshi Nagatsuka8Yoshihiko Furuki9Department of Oral and Maxillofacial Surgery, Kagawa Prefectural Central Hospital, 1-2-1, Asahi-machi, Takamatsu, Kagawa 760-8557, JapanElectronic and Computer Engineering, Department of Electrical, Faculty of Engineering, Gifu University, 1-1 Yanagido, Gifu, Gifu 501-1193, JapanElectronic and Computer Engineering, Department of Electrical, Faculty of Engineering, Gifu University, 1-1 Yanagido, Gifu, Gifu 501-1193, JapanDepartment of Oral and Maxillofacial Surgery, Kagawa Prefectural Central Hospital, 1-2-1, Asahi-machi, Takamatsu, Kagawa 760-8557, JapanPolytechnic Center Kagawa, 2-4-3, Hananomiya-cho, Takamatsu, Kagawa 761-8063, JapanDentistry and Pharmaceutical Sciences, Department of Oral Pathology and Medicine, Okayama University Graduate School of Medicine, Okayama 700-8558, JapanDentistry and Pharmaceutical Sciences, Department of Oral Pathology and Medicine, Okayama University Graduate School of Medicine, Okayama 700-8558, JapanDentistry and Pharmaceutical Sciences, Department of Oral Pathology and Medicine, Okayama University Graduate School of Medicine, Okayama 700-8558, JapanDentistry and Pharmaceutical Sciences, Department of Oral Pathology and Medicine, Okayama University Graduate School of Medicine, Okayama 700-8558, JapanDepartment of Oral and Maxillofacial Surgery, Kagawa Prefectural Central Hospital, 1-2-1, Asahi-machi, Takamatsu, Kagawa 760-8557, JapanIt is necessary to accurately identify dental implant brands and the stage of treatment to ensure efficient care. Thus, the purpose of this study was to use multi-task deep learning to investigate a classifier that categorizes implant brands and treatment stages from dental panoramic radiographic images. For objective labeling, 9767 dental implant images of 12 implant brands and treatment stages were obtained from the digital panoramic radiographs of patients who underwent procedures at Kagawa Prefectural Central Hospital, Japan, between 2005 and 2020. Five deep convolutional neural network (CNN) models (ResNet18, 34, 50, 101 and 152) were evaluated. The accuracy, precision, recall, specificity, F1 score, and area under the curve score were calculated for each CNN. We also compared the multi-task and single-task accuracies of brand classification and implant treatment stage classification. Our analysis revealed that the larger the number of parameters and the deeper the network, the better the performance for both classifications. Multi-tasking significantly improved brand classification on all performance indicators, except recall, and significantly improved all metrics in treatment phase classification. Using CNNs conferred high validity in the classification of dental implant brands and treatment stages. Furthermore, multi-task learning facilitated analysis accuracy.https://www.mdpi.com/2218-273X/11/6/815multi-task learningdeep learningartificial intelligencedental implantclassification
collection DOAJ
language English
format Article
sources DOAJ
author Shintaro Sukegawa
Kazumasa Yoshii
Takeshi Hara
Tamamo Matsuyama
Katsusuke Yamashita
Keisuke Nakano
Kiyofumi Takabatake
Hotaka Kawai
Hitoshi Nagatsuka
Yoshihiko Furuki
spellingShingle Shintaro Sukegawa
Kazumasa Yoshii
Takeshi Hara
Tamamo Matsuyama
Katsusuke Yamashita
Keisuke Nakano
Kiyofumi Takabatake
Hotaka Kawai
Hitoshi Nagatsuka
Yoshihiko Furuki
Multi-Task Deep Learning Model for Classification of Dental Implant Brand and Treatment Stage Using Dental Panoramic Radiograph Images
Biomolecules
multi-task learning
deep learning
artificial intelligence
dental implant
classification
author_facet Shintaro Sukegawa
Kazumasa Yoshii
Takeshi Hara
Tamamo Matsuyama
Katsusuke Yamashita
Keisuke Nakano
Kiyofumi Takabatake
Hotaka Kawai
Hitoshi Nagatsuka
Yoshihiko Furuki
author_sort Shintaro Sukegawa
title Multi-Task Deep Learning Model for Classification of Dental Implant Brand and Treatment Stage Using Dental Panoramic Radiograph Images
title_short Multi-Task Deep Learning Model for Classification of Dental Implant Brand and Treatment Stage Using Dental Panoramic Radiograph Images
title_full Multi-Task Deep Learning Model for Classification of Dental Implant Brand and Treatment Stage Using Dental Panoramic Radiograph Images
title_fullStr Multi-Task Deep Learning Model for Classification of Dental Implant Brand and Treatment Stage Using Dental Panoramic Radiograph Images
title_full_unstemmed Multi-Task Deep Learning Model for Classification of Dental Implant Brand and Treatment Stage Using Dental Panoramic Radiograph Images
title_sort multi-task deep learning model for classification of dental implant brand and treatment stage using dental panoramic radiograph images
publisher MDPI AG
series Biomolecules
issn 2218-273X
publishDate 2021-05-01
description It is necessary to accurately identify dental implant brands and the stage of treatment to ensure efficient care. Thus, the purpose of this study was to use multi-task deep learning to investigate a classifier that categorizes implant brands and treatment stages from dental panoramic radiographic images. For objective labeling, 9767 dental implant images of 12 implant brands and treatment stages were obtained from the digital panoramic radiographs of patients who underwent procedures at Kagawa Prefectural Central Hospital, Japan, between 2005 and 2020. Five deep convolutional neural network (CNN) models (ResNet18, 34, 50, 101 and 152) were evaluated. The accuracy, precision, recall, specificity, F1 score, and area under the curve score were calculated for each CNN. We also compared the multi-task and single-task accuracies of brand classification and implant treatment stage classification. Our analysis revealed that the larger the number of parameters and the deeper the network, the better the performance for both classifications. Multi-tasking significantly improved brand classification on all performance indicators, except recall, and significantly improved all metrics in treatment phase classification. Using CNNs conferred high validity in the classification of dental implant brands and treatment stages. Furthermore, multi-task learning facilitated analysis accuracy.
topic multi-task learning
deep learning
artificial intelligence
dental implant
classification
url https://www.mdpi.com/2218-273X/11/6/815
work_keys_str_mv AT shintarosukegawa multitaskdeeplearningmodelforclassificationofdentalimplantbrandandtreatmentstageusingdentalpanoramicradiographimages
AT kazumasayoshii multitaskdeeplearningmodelforclassificationofdentalimplantbrandandtreatmentstageusingdentalpanoramicradiographimages
AT takeshihara multitaskdeeplearningmodelforclassificationofdentalimplantbrandandtreatmentstageusingdentalpanoramicradiographimages
AT tamamomatsuyama multitaskdeeplearningmodelforclassificationofdentalimplantbrandandtreatmentstageusingdentalpanoramicradiographimages
AT katsusukeyamashita multitaskdeeplearningmodelforclassificationofdentalimplantbrandandtreatmentstageusingdentalpanoramicradiographimages
AT keisukenakano multitaskdeeplearningmodelforclassificationofdentalimplantbrandandtreatmentstageusingdentalpanoramicradiographimages
AT kiyofumitakabatake multitaskdeeplearningmodelforclassificationofdentalimplantbrandandtreatmentstageusingdentalpanoramicradiographimages
AT hotakakawai multitaskdeeplearningmodelforclassificationofdentalimplantbrandandtreatmentstageusingdentalpanoramicradiographimages
AT hitoshinagatsuka multitaskdeeplearningmodelforclassificationofdentalimplantbrandandtreatmentstageusingdentalpanoramicradiographimages
AT yoshihikofuruki multitaskdeeplearningmodelforclassificationofdentalimplantbrandandtreatmentstageusingdentalpanoramicradiographimages
_version_ 1721411825777508352