Faculty of Data Science, Shiga University, 1-1-1 Banba, Hikone, Shiga, 522-8222, Japan.
Department of Intelligence Science and Engineering, Graduate School of Natural Science and Technology, Gifu University, 1-1 Yanagido, Gifu, 501-1194, Japan.
Oral Radiol. 2021 Jan;37(1):13-19. doi: 10.1007/s11282-019-00418-w. Epub 2020 Jan 1.
Dental state plays an important role in forensic radiology in case of large scale disasters. However, dental information stored in dental clinics are not standardized or electronically filed in general. The purpose of this study is to develop a computerized system to detect and classify teeth in dental panoramic radiographs for automatic structured filing of the dental charts. It can also be used as a preprocessing step for computerized image analysis of dental diseases.
One hundred dental panoramic radiographs were employed for training and testing an object detection network using fourfold cross-validation method. The detected bounding boxes were then classified into four tooth types, including incisors, canines, premolars, and molars, and three tooth conditions, including nonmetal restored, partially restored, and completely restored, using classification network. Based on the visualization result, multisized image data were used for the double input layers of a convolutional neural network. The result was evaluated by the detection sensitivity, the number of false-positive detection, and classification accuracies.
The tooth detection sensitivity was 96.4% with 0.5 false positives per case. The classification accuracies for tooth types and tooth conditions were 93.2% and 98.0%. Using the double input layer network, 6 point increase in classification accuracy was achieved for the tooth types.
The proposed method can be useful in automatic filing of dental charts for forensic identification and preprocessing of dental disease prescreening purposes.
在大规模灾难中，口腔状况在法医放射学中起着重要作用。然而，一般来说，存储在牙科诊所中的牙科信息没有标准化或电子存档。本研究的目的是开发一种计算机系统，用于检测和分类牙科全景片中的牙齿，以便自动对牙科图表进行结构化归档。它也可以用作牙科疾病计算机图像分析的预处理步骤。
使用四折交叉验证方法，对 100 张牙科全景片进行了对象检测网络的训练和测试。然后使用分类网络将检测到的边界框分类为四种牙齿类型，包括切牙、尖牙、前磨牙和磨牙，以及三种牙齿状况，包括非金属修复、部分修复和完全修复。基于可视化结果，使用多尺寸图像数据作为卷积神经网络的双输入层。通过检测灵敏度、假阳性检测数量和分类准确率来评估结果。
牙齿检测灵敏度为 96.4%，每例假阳性为 0.5 个。牙齿类型和牙齿状况的分类准确率分别为 93.2%和 98.0%。使用双输入层网络，牙齿类型的分类准确率提高了 6 个百分点。
该方法可用于自动归档法医识别的牙科图表，也可用于牙科疾病初筛的预处理。
