深度学习在口腔医学影像中的应用与挑战  

Applications and challenges of deep learning in dental imaging

作　　者：赵阳 李俊诚 成博栋 牛娜君 王龙光 高广谓 施俊[1] Zhao Yang;Li Juncheng;Cheng Bodong;Niu Najun;Wang Longguang;Gao Guangwei;Shi Jun(School of Communication and Information Engineering,Shanghai University,Shanghai 200444,China;School of Computer Science and Technology,Xidian University,Xi’an 710000,China;School of Stomatology,Nanjing Medical University,Nanjing 210000,China;College of Electronic Science,Air Force Aviation University,Jilin 130022,China;Institute of Advanced Technology,Nanjing University of Posts and Telecommunications,Nanjing 210003,China)

机构地区：[1]上海大学通信与信息工程学院,上海200444 [2]西安电子科技大学计算机科学与技术学院,西安710000 [3]南京医科大学口腔医学院,南京210000 [4]空军航空大学电子科学学院,吉林130022 [5]南京邮电大学先进技术研究院,南京210003

出　　处：《中国图象图形学报》2024年第3期586-607,共22页Journal of Image and Graphics

基　　金：上海市自然科学基金项目(23ZR1422200);上海市扬帆计划项目(23YF1412800)。

摘　　要：口腔医学影像是进行临床口腔疾病检测、筛查、诊断和治疗评估的重要工具,对口腔影像进行准确分析对于后续治疗计划的制定至关重要。常规的口腔医学影像分析依赖于医师的水平和经验,存在阅片效率低、可重复性低以及定量分析欠缺的问题。深度学习可以从大样本数据中自动学习并获取优良的特征表达,提升各类机器学习任务的效率和性能,目前已广泛应用于医学影像分析处理的各类任务之中。基于深度学习的口腔医学影像处理是目前的研究热点,但由于口腔医学领域内在的特殊性和复杂性,以及口腔医学影像数据样本量通常较小的问题,给深度学习方法在相关学习任务和场景的应用带来了新的挑战。本文从口腔医学影像领域常用的二维X射线影像、三维点云/网格影像和锥形束计算机断层扫描影像3种影像出发,介绍深度学习技术在口腔医学影像处理及分析领域应用的思路和现状,分析了各算法的优缺点及该领域所面临的问题和挑战,并对未来的研究方向和可能开展的临床应用进行展望,以助力智慧口腔建设。Dental imaging is an essential tool for the detection,screening,diagnosis,and therapeutic evaluation of clinical oral diseases,and the accurate analysis of the images is vital to the development of subsequent treatment plans.Deep learn⁃ing,which is widely used in many fields,such as machine translation,speech recognition,and computer vision,can auto⁃matically learn and obtain superior feature expressions from large sample data,thereby improving the efficiency and perfor⁃mance of various machine learning tasks.With the integration of artificial intelligence and various fields,smart healthcare has become an important application area of deep learning,providing an effective way of solving the following clinical prob⁃lems:1)the shortage of experienced radiologists in the field of dentistry cannot meet the rapidly growing medical demand.2)Despite sufficient medical resources,the number of experienced physicians cannot meet the rapidly growing medical demand.3)Different physicians have different interpretations of the same oral image,which are influenced by subjectiv⁃ity.Deep learning-based dental image processing is currently a popular research topic.The inherent specificity and com⁃plexity of the medical field and the problem of insufficient dental image data samples bring new challenges to the applica⁃tion of deep learning methods in relevant learning tasks and scenarios.This work mainly reviews the various applications of deep learning methods using three major dental imaging methods(i.e.,two-dimensional oral X-ray images,threedimensional tooth point cloud/mesh images,cone beam computed tomography(CBCT)).These applications include tooth segmentation,caries detection,and tumor detection.The reviews on two-dimensional oral X-ray images focus on bitewing,periapical,and panoramic X-rays based on deep learning methods.Bitewing X-rays usually show the contact surface from the distal end of the canine to the most distal molar and are mainly used to diagnose proximal caries,assess the extent of caries,and ide

关 键 词：深度学习 口腔医学影像 牙齿检测与分割 龋齿检测 计算机辅助诊断(CAD) 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]