Hyun-Jun Kong, Sang-Ho Eom, Jin-Yong Yoo, Jun-Hyeok Lee
{"title":"用集成深度学习识别130种牙种植体类型。","authors":"Hyun-Jun Kong, Sang-Ho Eom, Jin-Yong Yoo, Jun-Hyeok Lee","doi":"10.11607/jomi.9818","DOIUrl":null,"url":null,"abstract":"<p><p><b>Purpose:</b> To evaluate the accuracy and clinical usability of an identification model using ensemble deep learning for 130 dental implant types. <b>Materials and Methods:</b> A total of 28,112 panoramic radiographs were obtained from 30 domestic and foreign dental clinics. From these panoramic radiographs, 45,909 implant fixture images were extracted and labeled based on electronic medical records. Dental implants were classified into 130 types according to the manufacturer, the manufacturer's implant system, and the diameter and length of the implant fixture. Regions of interest were manually cropped, and data augmentation was performed. According to the minimum number of images collected per implant type, the datasets were classified into three sets: an overall total of 130 and two subsets that consisted of 79 and 58 types. EfficientNet and Res2Next algorithms were used for image classification in deep learning. After testing the performance of the two models, the ensemble learning technique was applied to improve accuracy. The top-1 accuracy, top-5 accuracy, precision, recall, and F1 scores were calculated according to algorithms and datasets. <b>Results:</b> For the 130 types, the top-1 accuracy, top-5 accuracy, precision, recall, and F1 scores were 75.27, 95.02, 78.84, 75.27, and 74.89, respectively. In all cases, the ensemble model performed better than EfficientNet and Res2Next. When using the ensemble model, the accuracy increased as the number of types decreased. <b>Conclusion:</b> The ensemble deep learning model for the identification of 130 types of dental implants showed higher accuracy than the existing algorithms. To further improve the performance and clinical usability of the model, images with higher quality and fine-tuned algorithms optimized for implant identification are required.</p>","PeriodicalId":50298,"journal":{"name":"International Journal of Oral & Maxillofacial Implants","volume":"38 1","pages":"150-156"},"PeriodicalIF":1.7000,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":"{\"title\":\"Identification of 130 Dental Implant Types Using Ensemble Deep Learning.\",\"authors\":\"Hyun-Jun Kong, Sang-Ho Eom, Jin-Yong Yoo, Jun-Hyeok Lee\",\"doi\":\"10.11607/jomi.9818\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><p><b>Purpose:</b> To evaluate the accuracy and clinical usability of an identification model using ensemble deep learning for 130 dental implant types. <b>Materials and Methods:</b> A total of 28,112 panoramic radiographs were obtained from 30 domestic and foreign dental clinics. From these panoramic radiographs, 45,909 implant fixture images were extracted and labeled based on electronic medical records. Dental implants were classified into 130 types according to the manufacturer, the manufacturer's implant system, and the diameter and length of the implant fixture. Regions of interest were manually cropped, and data augmentation was performed. According to the minimum number of images collected per implant type, the datasets were classified into three sets: an overall total of 130 and two subsets that consisted of 79 and 58 types. EfficientNet and Res2Next algorithms were used for image classification in deep learning. After testing the performance of the two models, the ensemble learning technique was applied to improve accuracy. The top-1 accuracy, top-5 accuracy, precision, recall, and F1 scores were calculated according to algorithms and datasets. <b>Results:</b> For the 130 types, the top-1 accuracy, top-5 accuracy, precision, recall, and F1 scores were 75.27, 95.02, 78.84, 75.27, and 74.89, respectively. In all cases, the ensemble model performed better than EfficientNet and Res2Next. When using the ensemble model, the accuracy increased as the number of types decreased. <b>Conclusion:</b> The ensemble deep learning model for the identification of 130 types of dental implants showed higher accuracy than the existing algorithms. To further improve the performance and clinical usability of the model, images with higher quality and fine-tuned algorithms optimized for implant identification are required.</p>\",\"PeriodicalId\":50298,\"journal\":{\"name\":\"International Journal of Oral & Maxillofacial Implants\",\"volume\":\"38 1\",\"pages\":\"150-156\"},\"PeriodicalIF\":1.7000,\"publicationDate\":\"2023-01-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"3\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"International Journal of Oral & Maxillofacial Implants\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://doi.org/10.11607/jomi.9818\",\"RegionNum\":4,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q3\",\"JCRName\":\"DENTISTRY, ORAL SURGERY & MEDICINE\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Journal of Oral & Maxillofacial Implants","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.11607/jomi.9818","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"DENTISTRY, ORAL SURGERY & MEDICINE","Score":null,"Total":0}
Identification of 130 Dental Implant Types Using Ensemble Deep Learning.
Purpose: To evaluate the accuracy and clinical usability of an identification model using ensemble deep learning for 130 dental implant types. Materials and Methods: A total of 28,112 panoramic radiographs were obtained from 30 domestic and foreign dental clinics. From these panoramic radiographs, 45,909 implant fixture images were extracted and labeled based on electronic medical records. Dental implants were classified into 130 types according to the manufacturer, the manufacturer's implant system, and the diameter and length of the implant fixture. Regions of interest were manually cropped, and data augmentation was performed. According to the minimum number of images collected per implant type, the datasets were classified into three sets: an overall total of 130 and two subsets that consisted of 79 and 58 types. EfficientNet and Res2Next algorithms were used for image classification in deep learning. After testing the performance of the two models, the ensemble learning technique was applied to improve accuracy. The top-1 accuracy, top-5 accuracy, precision, recall, and F1 scores were calculated according to algorithms and datasets. Results: For the 130 types, the top-1 accuracy, top-5 accuracy, precision, recall, and F1 scores were 75.27, 95.02, 78.84, 75.27, and 74.89, respectively. In all cases, the ensemble model performed better than EfficientNet and Res2Next. When using the ensemble model, the accuracy increased as the number of types decreased. Conclusion: The ensemble deep learning model for the identification of 130 types of dental implants showed higher accuracy than the existing algorithms. To further improve the performance and clinical usability of the model, images with higher quality and fine-tuned algorithms optimized for implant identification are required.
期刊介绍:
Edited by Steven E. Eckert, DDS, MS ISSN (Print): 0882-2786
ISSN (Online): 1942-4434
This highly regarded, often-cited journal integrates clinical and scientific data to improve methods and results of oral and maxillofacial implant therapy. It presents pioneering research, technology, clinical applications, reviews of the literature, seminal studies, emerging technology, position papers, and consensus studies, as well as the many clinical and therapeutic innovations that ensue as a result of these efforts. The editorial board is composed of recognized opinion leaders in their respective areas of expertise and reflects the international reach of the journal. Under their leadership, JOMI maintains its strong scientific integrity while expanding its influence within the field of implant dentistry. JOMI’s popular regular feature "Thematic Abstract Review" presents a review of abstracts of recently published articles on a specific topical area of interest each issue.