Eichner classification based on panoramic X-ray images using deep learning: A pilot study.

IF 1 4区医学 Q4 ENGINEERING, BIOMEDICAL

Bio-medical materials and engineering Pub Date : 2024-01-01 DOI:10.3233/BME-230217

Yuta Otsuka, Hiroko Indo, Yusuke Kawashima, Tatsuro Tanaka, Hiroshi Kono, Masafumi Kikuchi

{"title":"Eichner classification based on panoramic X-ray images using deep learning: A pilot study.","authors":"Yuta Otsuka, Hiroko Indo, Yusuke Kawashima, Tatsuro Tanaka, Hiroshi Kono, Masafumi Kikuchi","doi":"10.3233/BME-230217","DOIUrl":null,"url":null,"abstract":"Background: Research using panoramic X-ray images using deep learning has been progressing in recent years. There is a need to propose methods that can classify and predict from image information.Objective: In this study, Eichner classification was performed on image processing based on panoramic X-ray images. The Eichner classification was based on the remaining teeth, with the aim of making partial dentures. This classification was based on the condition that the occlusal position was supported by the remaining teeth in the upper and lower jaws.Methods: Classification models were constructed using two convolutional neural network methods: the sequential and VGG19 models. The accuracy was compared with the accuracy of Eichner classification using the sequential and VGG19 models.Results: Both accuracies were greater than 81%, and they had sufficient functions for the Eichner classification.Conclusion: We were able to build a highly accurate prediction model using deep learning scratch sequential model and VGG19. This predictive model will become part of the basic considerations for future AI research in dentistry.","PeriodicalId":9109,"journal":{"name":"Bio-medical materials and engineering","volume":" ","pages":"377-386"},"PeriodicalIF":1.0000,"publicationDate":"2024-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Bio-medical materials and engineering","FirstCategoryId":"5","ListUrlMain":"https://doi.org/10.3233/BME-230217","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"ENGINEERING, BIOMEDICAL","Score":null,"Total":0}

引用次数: 0

Abstract

Background: Research using panoramic X-ray images using deep learning has been progressing in recent years. There is a need to propose methods that can classify and predict from image information.

Objective: In this study, Eichner classification was performed on image processing based on panoramic X-ray images. The Eichner classification was based on the remaining teeth, with the aim of making partial dentures. This classification was based on the condition that the occlusal position was supported by the remaining teeth in the upper and lower jaws.

Methods: Classification models were constructed using two convolutional neural network methods: the sequential and VGG19 models. The accuracy was compared with the accuracy of Eichner classification using the sequential and VGG19 models.

Results: Both accuracies were greater than 81%, and they had sufficient functions for the Eichner classification.

Conclusion: We were able to build a highly accurate prediction model using deep learning scratch sequential model and VGG19. This predictive model will become part of the basic considerations for future AI research in dentistry.

查看原文本刊更多论文

利用深度学习对全景 X 光图像进行艾希纳分类：试点研究。

背景：近年来，利用深度学习对全景 X 光图像进行的研究不断取得进展。有必要提出能根据图像信息进行分类和预测的方法：本研究基于全景 X 光图像，对图像处理进行了艾希纳分类。艾希纳分类法以剩余牙齿为基础，目的是制作局部义齿。这种分类是基于咬合位置由上下颌剩余牙齿支撑这一条件：使用两种卷积神经网络方法构建了分类模型：序列模型和 VGG19 模型。结果：两个模型的准确率均高于 Eichner 分类法：结果：两种模型的准确率都超过了 81%，而且它们对艾希纳分类法具有足够的功能：结论：我们利用深度学习划痕序列模型和 VGG19 建立了一个高精度的预测模型。该预测模型将成为未来牙科人工智能研究的基本考虑因素之一。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Bio-medical materials and engineering 工程技术-材料科学：生物材料

CiteScore

1.80

自引率

0.00%

发文量

审稿时长

6 months

期刊介绍： The aim of Bio-Medical Materials and Engineering is to promote the welfare of humans and to help them keep healthy. This international journal is an interdisciplinary journal that publishes original research papers, review articles and brief notes on materials and engineering for biological and medical systems. Articles in this peer-reviewed journal cover a wide range of topics, including, but not limited to: Engineering as applied to improving diagnosis, therapy, and prevention of disease and injury, and better substitutes for damaged or disabled human organs; Studies of biomaterial interactions with the human body, bio-compatibility, interfacial and interaction problems; Biomechanical behavior under biological and/or medical conditions; Mechanical and biological properties of membrane biomaterials; Cellular and tissue engineering, physiological, biophysical, biochemical bioengineering aspects; Implant failure fields and degradation of implants. Biomimetics engineering and materials including system analysis as supporter for aged people and as rehabilitation; Bioengineering and materials technology as applied to the decontamination against environmental problems; Biosensors, bioreactors, bioprocess instrumentation and control system; Application to food engineering; Standardization problems on biomaterials and related products; Assessment of reliability and safety of biomedical materials and man-machine systems; and Product liability of biomaterials and related products.