Deep Learning-Based Detection of Impacted Teeth on Panoramic Radiographs.

IF 2.3 Q3 ENGINEERING, BIOMEDICAL

Biomedical Engineering and Computational Biology Pub Date : 2024-10-05 eCollection Date: 2024-01-01 DOI:10.1177/11795972241288319

He Zhicheng, Wang Yipeng, Li Xiao

{"title":"Deep Learning-Based Detection of Impacted Teeth on Panoramic Radiographs.","authors":"He Zhicheng, Wang Yipeng, Li Xiao","doi":"10.1177/11795972241288319","DOIUrl":null,"url":null,"abstract":"Objective: The aim is to detect impacted teeth in panoramic radiology by refining the pretrained MedSAM model.Study design: Impacted teeth are dental issues that can cause complications and are diagnosed via radiographs. We modified SAM model for individual tooth segmentation using 1016 X-ray images. The dataset was split into training, validation, and testing sets, with a ratio of 16:3:1. We enhanced the SAM model to automatically detect impacted teeth by focusing on the tooth's center for more accurate results.Results: With 200 epochs, batch size equals to 1, and a learning rate of 0.001, random images trained the model. Results on the test set showcased performance up to an accuracy of 86.73%, F1-score of 0.5350, and IoU of 0.3652 on SAM-related models.Conclusion: This study fine-tunes MedSAM for impacted tooth segmentation in X-ray images, aiding dental diagnoses. Further improvements on model accuracy and selection are essential for enhancing dental practitioners' diagnostic capabilities.","PeriodicalId":42484,"journal":{"name":"Biomedical Engineering and Computational Biology","volume":null,"pages":null},"PeriodicalIF":2.3000,"publicationDate":"2024-10-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11456186/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Biomedical Engineering and Computational Biology","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1177/11795972241288319","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2024/1/1 0:00:00","PubModel":"eCollection","JCR":"Q3","JCRName":"ENGINEERING, BIOMEDICAL","Score":null,"Total":0}

引用次数: 0

Abstract

Objective: The aim is to detect impacted teeth in panoramic radiology by refining the pretrained MedSAM model.

Study design: Impacted teeth are dental issues that can cause complications and are diagnosed via radiographs. We modified SAM model for individual tooth segmentation using 1016 X-ray images. The dataset was split into training, validation, and testing sets, with a ratio of 16:3:1. We enhanced the SAM model to automatically detect impacted teeth by focusing on the tooth's center for more accurate results.

Results: With 200 epochs, batch size equals to 1, and a learning rate of 0.001, random images trained the model. Results on the test set showcased performance up to an accuracy of 86.73%, F1-score of 0.5350, and IoU of 0.3652 on SAM-related models.

Conclusion: This study fine-tunes MedSAM for impacted tooth segmentation in X-ray images, aiding dental diagnoses. Further improvements on model accuracy and selection are essential for enhancing dental practitioners' diagnostic capabilities.

查看原文本刊更多论文

基于深度学习的全景 X 光片牙齿撞击检测。

研究目的研究设计：研究设计：撞击牙是一种可引起并发症的牙科问题，可通过 X 光片进行诊断。我们利用 1016 张 X 光图像修改了用于单个牙齿分割的 SAM 模型。数据集分为训练集、验证集和测试集，比例为 16:3:1。我们对 SAM 模型进行了改进，通过聚焦牙齿中心来自动检测撞击牙齿，从而获得更准确的结果：在 200 个历元、批量大小等于 1 和学习率为 0.001 的条件下，随机图像对模型进行了训练。测试集的结果显示，SAM 相关模型的准确率高达 86.73%，F1 分数为 0.5350，IoU 为 0.3652：本研究对 MedSAM 进行了微调，用于 X 射线图像中的撞击牙分割，为牙科诊断提供了帮助。要提高牙科医生的诊断能力，进一步提高模型的准确性和选择至关重要。

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

Biomedical Engineering and Computational Biology ENGINEERING, BIOMEDICAL-

自引率

0.00%

发文量

审稿时长

8 weeks