帮助 关于我们

返回检索结果

深度学习在基于曲面体层片的成釉细胞瘤及牙源性角化囊肿鉴别诊断中的应用
Application of Deep Learning in Differential Diagnosis of Ameloblastoma and Odontogenic Keratocyst Based on Panoramic Radiographs

查看参考文献22篇

李敏 1,2   慕创创 1,2   张建运 2,3   李刚 1,2 *  
文摘 目的通过应用不同卷积神经网络模型对成釉细胞瘤及牙源性角化囊肿进行鉴别诊断。方法回顾性收集1000张成釉细胞瘤和牙源性角化囊肿患者的数字曲面体层片,选用ResNet(18、50、101)、VGG(16、19)、EfficientNet(b1、 b3、b5)深度学习模型,对训练集中的800张曲面体层片经五折交叉验证的方法训练后对测试集中的200张曲面体层片进行鉴别诊断。同时, 7名口腔放射专业医生对这200张曲面体层片进行诊断,并将二者的诊断结果进行比较。结果卷积神经网络模型的诊断准确率为82.50%~87.50%,其中EfficientNet b1准确率最高,为87.50%,各卷积神经网络模型训练集和测试集本身之间比较,准确率差异无统计学意义(P_(训练集)=0.998, P_(测试集)=0.905)。7名口腔放射专业医生(2名高年资医生、 5名低年资医生)平均诊断准确率为(70.30 ± 5.48)%,且不同年资医生之间平均诊断准确率差异无统计学意义(P = 0.883)。深度学习卷积神经网络模型的诊断准确率显著高于口腔放射专业医生的诊断准确率(P <0.001)。结论基于曲面体层片的深度学习卷积神经网络能够对成釉细胞瘤和牙源性角化囊肿做出较为准确的鉴别诊断。
其他语种文摘 Objective To evaluate the accuracy of different convolutional neural networks(CNN), representative deep learning models, in the differential diagnosis of ameloblastoma and odontogenic keratocyst, and subsequently compare the diagnosis results between models and oral radiologists. Methods A total of 1000 digital panoramic radiographs were retrospectively collected from the patients with ameloblastoma(500 radiographs)or odontogenic keratocyst(500 radiographs)in the Department of Oral and Maxillofacial Radiology, Peking University School of Stomatology. Eight CNN including ResNet(18, 50, 101), VGG(16, 19), and EfficientNet(b1, b3, b5)were selected to distinguish ameloblastoma from odontogenic keratocyst. Transfer learning was employed to train 800 panoramic radiographs in the training set through 5-fold cross validation, and 200 panoramic radiographs in the test set were used for differential diagnosis. Chi square test was performed for comparing the performance among different CNN. Furthermore, 7 oral radiologists(including 2 seniors and 5 juniors)made a diagnosis on the 200 panoramic radiographs in the test set, and the diagnosis results were compared between CNN and oral radiologists. Results The eight neural network models showed the diagnostic accuracy ranging from 82.50% to 87.50%, of which EfficientNet b1 had the highest accuracy of 87.50%. There was no significant difference in the diagnostic accuracy among the CNN models(P =0.998, P =0.905). The average diagnostic accuracy of oral radiologists was(70.30 ±5.48)%, and there was no statistical difference in the accuracy between senior and junior oral radiologists(P =0.883). The diagnostic accuracy of CNN models was higher than that of oral radiologists(P <0.001). Conclusion Deep learning CNN can realize accurate differential diagnosis between ameloblastoma and odontogenic keratocyst with panoramic radiographs, with higher diagnostic accuracy than oral radiologists.
来源 中国医学科学院学报 ,2023,45(2):273-279 【核心库】
DOI 10.3881/j.issn.1000-503X.15139
关键词 卷积神经网络 ; 深度学习 ; 曲面体层片 ; 成釉细胞瘤 ; 牙源性角化囊肿
地址

1. 北京大学口腔医学院口腔医院医学影像科, 北京, 100081  

2. 国家口腔疾病临床医学研究中心口腔生物材料和数字诊疗装备国家工程研究中心口腔数字医学北京市重点实验室, 国家口腔疾病临床医学研究中心;;口腔生物材料和数字诊疗装备国家工程研究中心;;口腔数字医学北京市重点实验室, 北京, 100081  

3. 北京大学口腔医学院口腔医院医学病理科, 北京, 100081

语种 中文
文献类型 研究性论文
ISSN 1000-503X
学科 肿瘤学
基金 北京大学百度基金
文献收藏号 CSCD:7470004

参考文献 共 22 共2页

1.  郭传瑸. 口腔颌面外科学. (3版),2021 CSCD被引 1 次    
2.  Ariji Y. Imaging features contributing to the diagnosis of ameloblastomas and keratocystic odontogenic tumours: logistic regression analysis. Dentomaxillofac Radiol,2011,40(3):133-140 CSCD被引 7 次    
3.  Chai Z K. Improved diagnostic accuracy of ameloblastoma and odontogenic keratocyst on conebeam CT by artificial intelligence. Front Oncol,2022,11:793417 CSCD被引 2 次    
4.  Feng Y. Breast cancer cell nuclei classification in histopathology images using deep neural networks. Int J Comput Assist Radiol Surg,2018,13(2):179-191 CSCD被引 2 次    
5.  van Ginneken B. Fifty years of computer analysis in chest imaging: rule-based, machine learning, deep learning. Radiol Phys Technol,2017,10(1):23-32 CSCD被引 12 次    
6.  Ren S. Faster R-CNN: towards real-time object detection with region proposal networks. IEEE Trans Pattern Anal Mach Intell,2017,39(6):1137-1149 CSCD被引 4396 次    
7.  Shichijo S. Application of convolutional neural networks in the diagnosis of helicobacter pylori infection based on endoscopic images. EBioMedicine,2017,25:106-111 CSCD被引 23 次    
8.  Du X. A convolutional neural network based auto-positioning method for dental arch in rotational panoramic radiography. Conf Proc IEEE Eng Med Biol Soc. 2018,2018:2615-2618 CSCD被引 1 次    
9.  Lee J H. Detection and diagnosis of dental caries using a deep learning-based convolutional neural network algorithm. J Dent,2018,77:106-111 CSCD被引 13 次    
10.  Ekert T. Deep learning for the radiographic detection of apical lesions. J Endod,2019,45(7):917-922.e5 CSCD被引 7 次    
11.  Farhadian M. Dental age estimation using the pulp-to-tooth ratio in canines by neural networks. Imaging Sci Dent,2019,49(1):19-26 CSCD被引 2 次    
12.  Vila-Blanco N. Deep neural networks for chronological age estimation from OPG images. IEEE Trans Med Imaging,2020,39(7):2374-2384 CSCD被引 2 次    
13.  马绪臣. 口腔颌面影像学. (2版),2014 CSCD被引 1 次    
14.  张祖燕. 口腔颌面医学影像诊断学. (7版),2020 CSCD被引 4 次    
15.  Minami M. Cystic lesions of the maxillomandibular region: MR imaging distinction of odontogenic keratocysts and ameloblastomas from other cysts. AJR Am J Roentgenol,1996,166(4):943-949 CSCD被引 5 次    
16.  Han Y. Diffusion-weighted MR imaging of unicystic odontogenic tumors for differentiation of unicystic ameloblastomas from keratocystic odontogenic tumors. Korean J Radiol,2018,19(1):79-84 CSCD被引 1 次    
17.  Fujita M. Diagnostic value of MRI for odontogenic tumours. Dentomaxillofac Radiol,2013,42(5):20120265 CSCD被引 2 次    
18.  Sugianto I. Diagnostic value of fluid attenuated inversion recovery magnetic resonance imaging for multilocular ameloblastoma. J Hard Tissue Biol,2018,27(4):275-280 CSCD被引 1 次    
19.  Greenspan H. Guest editorial deep learning in medical imaging: overview and future promise of an exciting new technique. IEEE Trans Med Imaging,2016,35(5):1153-1159 CSCD被引 47 次    
20.  Poedjiastoeti W. Application of convolutional neural network in the diagnosis of jaw tumors. Healthc Inform Res,2018,24(3):236-241 CSCD被引 4 次    
引证文献 1 篇

1 赖丹琳 不同层数的Mask-RCNN模型自动检测成釉细胞瘤效能的比较 口腔医学研究,2023,39(12):1092-1096
CSCD被引 1 次

显示所有1篇文献

论文科学数据集
PlumX Metrics
相关文献

 作者相关
 关键词相关
 参考文献相关

版权所有 ©2008 中国科学院文献情报中心 制作维护：中国科学院文献情报中心
地址：北京中关村北四环西路33号 邮政编码：100190 联系电话：(010)82627496 E-mail:cscd@mail.las.ac.cn 京ICP备05002861号-4 | 京公网安备11010802043238号