Wei Liu, Xiang Li, Chang Liu, Ge Gao, Yutao Xiong, Tao Zhu, Wei Zeng, Jixiang Guo, Wei Tang
{"title":"利用深度学习对锥形束计算机断层扫描中的多类颌骨病变进行自动分类和分割。","authors":"Wei Liu, Xiang Li, Chang Liu, Ge Gao, Yutao Xiong, Tao Zhu, Wei Zeng, Jixiang Guo, Wei Tang","doi":"10.1093/dmfr/twae028","DOIUrl":null,"url":null,"abstract":"<p><strong>Objectives: </strong>To develop and validate a modified deep learning (DL) model based on nnU-Net for classifying and segmenting five-class jaw lesions using cone-beam CT (CBCT).</p><p><strong>Methods: </strong>A total of 368 CBCT scans (37 168 slices) were used to train a multi-class segmentation model. The data underwent manual annotation by two oral and maxillofacial surgeons (OMSs) to serve as ground truth. Sensitivity, specificity, precision, F1-score, and accuracy were used to evaluate the classification ability of the model and doctors, with or without artificial intelligence assistance. The dice similarity coefficient (DSC), average symmetric surface distance (ASSD), and segmentation time were used to evaluate the segmentation effect of the model.</p><p><strong>Results: </strong>The model achieved the dual task of classifying and segmenting jaw lesions in CBCT. For classification, the sensitivity, specificity, precision, and accuracy of the model were 0.871, 0.974, 0.874, and 0.891, respectively, surpassing oral and maxillofacial radiologists (OMFRs) and OMSs, approaching the specialist. With the model's assistance, the classification performance of OMFRs and OMSs improved, particularly for odontogenic keratocyst (OKC) and ameloblastoma (AM), with F1-score improvements ranging from 6.2% to 12.7%. For segmentation, the DSC was 87.2% and the ASSD was 1.359 mm. The model's average segmentation time was 40 ± 9.9 s, contrasting with 25 ± 7.2 min for OMSs.</p><p><strong>Conclusions: </strong>The proposed DL model accurately and efficiently classified and segmented five classes of jaw lesions using CBCT. In addition, it could assist doctors in improving classification accuracy and segmentation efficiency, particularly in distinguishing confusing lesions (eg, AM and OKC).</p>","PeriodicalId":11261,"journal":{"name":"Dento maxillo facial radiology","volume":null,"pages":null},"PeriodicalIF":2.9000,"publicationDate":"2024-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Automatic classification and segmentation of multiclass jaw lesions in cone-beam CT using deep learning.\",\"authors\":\"Wei Liu, Xiang Li, Chang Liu, Ge Gao, Yutao Xiong, Tao Zhu, Wei Zeng, Jixiang Guo, Wei Tang\",\"doi\":\"10.1093/dmfr/twae028\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><strong>Objectives: </strong>To develop and validate a modified deep learning (DL) model based on nnU-Net for classifying and segmenting five-class jaw lesions using cone-beam CT (CBCT).</p><p><strong>Methods: </strong>A total of 368 CBCT scans (37 168 slices) were used to train a multi-class segmentation model. The data underwent manual annotation by two oral and maxillofacial surgeons (OMSs) to serve as ground truth. Sensitivity, specificity, precision, F1-score, and accuracy were used to evaluate the classification ability of the model and doctors, with or without artificial intelligence assistance. The dice similarity coefficient (DSC), average symmetric surface distance (ASSD), and segmentation time were used to evaluate the segmentation effect of the model.</p><p><strong>Results: </strong>The model achieved the dual task of classifying and segmenting jaw lesions in CBCT. For classification, the sensitivity, specificity, precision, and accuracy of the model were 0.871, 0.974, 0.874, and 0.891, respectively, surpassing oral and maxillofacial radiologists (OMFRs) and OMSs, approaching the specialist. With the model's assistance, the classification performance of OMFRs and OMSs improved, particularly for odontogenic keratocyst (OKC) and ameloblastoma (AM), with F1-score improvements ranging from 6.2% to 12.7%. For segmentation, the DSC was 87.2% and the ASSD was 1.359 mm. The model's average segmentation time was 40 ± 9.9 s, contrasting with 25 ± 7.2 min for OMSs.</p><p><strong>Conclusions: </strong>The proposed DL model accurately and efficiently classified and segmented five classes of jaw lesions using CBCT. In addition, it could assist doctors in improving classification accuracy and segmentation efficiency, particularly in distinguishing confusing lesions (eg, AM and OKC).</p>\",\"PeriodicalId\":11261,\"journal\":{\"name\":\"Dento maxillo facial radiology\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":2.9000,\"publicationDate\":\"2024-10-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Dento maxillo facial radiology\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://doi.org/10.1093/dmfr/twae028\",\"RegionNum\":2,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"DENTISTRY, ORAL SURGERY & MEDICINE\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Dento maxillo facial radiology","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1093/dmfr/twae028","RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"DENTISTRY, ORAL SURGERY & MEDICINE","Score":null,"Total":0}
Automatic classification and segmentation of multiclass jaw lesions in cone-beam CT using deep learning.
Objectives: To develop and validate a modified deep learning (DL) model based on nnU-Net for classifying and segmenting five-class jaw lesions using cone-beam CT (CBCT).
Methods: A total of 368 CBCT scans (37 168 slices) were used to train a multi-class segmentation model. The data underwent manual annotation by two oral and maxillofacial surgeons (OMSs) to serve as ground truth. Sensitivity, specificity, precision, F1-score, and accuracy were used to evaluate the classification ability of the model and doctors, with or without artificial intelligence assistance. The dice similarity coefficient (DSC), average symmetric surface distance (ASSD), and segmentation time were used to evaluate the segmentation effect of the model.
Results: The model achieved the dual task of classifying and segmenting jaw lesions in CBCT. For classification, the sensitivity, specificity, precision, and accuracy of the model were 0.871, 0.974, 0.874, and 0.891, respectively, surpassing oral and maxillofacial radiologists (OMFRs) and OMSs, approaching the specialist. With the model's assistance, the classification performance of OMFRs and OMSs improved, particularly for odontogenic keratocyst (OKC) and ameloblastoma (AM), with F1-score improvements ranging from 6.2% to 12.7%. For segmentation, the DSC was 87.2% and the ASSD was 1.359 mm. The model's average segmentation time was 40 ± 9.9 s, contrasting with 25 ± 7.2 min for OMSs.
Conclusions: The proposed DL model accurately and efficiently classified and segmented five classes of jaw lesions using CBCT. In addition, it could assist doctors in improving classification accuracy and segmentation efficiency, particularly in distinguishing confusing lesions (eg, AM and OKC).
期刊介绍:
Dentomaxillofacial Radiology (DMFR) is the journal of the International Association of Dentomaxillofacial Radiology (IADMFR) and covers the closely related fields of oral radiology and head and neck imaging.
Established in 1972, DMFR is a key resource keeping dentists, radiologists and clinicians and scientists with an interest in Head and Neck imaging abreast of important research and developments in oral and maxillofacial radiology.
The DMFR editorial board features a panel of international experts including Editor-in-Chief Professor Ralf Schulze. Our editorial board provide their expertise and guidance in shaping the content and direction of the journal.
Quick Facts:
- 2015 Impact Factor - 1.919
- Receipt to first decision - average of 3 weeks
- Acceptance to online publication - average of 3 weeks
- Open access option
- ISSN: 0250-832X
- eISSN: 1476-542X