Pan Xiao, Wang Chengtao, Luo Xuhui, Dong Qi, Sun Haiyang, Zhang Wentao, Qu Hongyan, Deng Runzhi, Lin Zitong
Department of Dentomaxillofacial Radiology, Nanjing Stomatological Hospital, Affiliated Hospital of Medical School, Institute of Stomatology, Nanjing University, Zhong Yang Road 30, Nanjing, Jiangsu Province, 210008, China.
Department of Preventive Dentistry, Nanjing Stomatological Hospital, Affiliated Hospital of Medical School, Institute of Stomatology, Nanjing University, Nanjing, China.
BMC Oral Health. 2025 Aug 21;25(1):1352. doi: 10.1186/s12903-025-06724-6.
Development and verification of a convolutional neural network (CNN)-based deep learning (DL) model for mandibular canal (MC) localization on multicenter cone beam computed tomography (CBCT) images.
In this study, a total 1056 CBCT scans in multiple centers were collected. Of these, 836 CBCT scans of one manufacturer were used for development of CNN model (training set: validation set: internal testing set = 640:360:36) and an external testing dataset of 220 CBCT scans from other four manufacturers were tested. The convolution module was built using a stack of Conv + InstanceNorm + LeakyReLU. Average symmetric surface distance (ASSD) and symmetric mean curve distance (SMCD) were used for quantitative evaluation of this model for both internal testing data and partial external testing data. Visual scoring (1-5 points) were performed to evaluate the accuracy and generalizability of MC localization for all external testing data. The differences of ASSD, SMCD and visual scores among the four manufacturers were compared for external testing dataset. The time of manual and automatic MC localization were recorded.
For the internal testing dataset, the average ASSD and SMCD was 0.486 mm and 0.298 mm respectively. For the external testing dataset, 86.8% CBCT scans' visual scores ≥ 4 points; the average ASSD and SMCD of 40 CBCT scans with visual scores ≥ 4 points were 0.438 mm and 0.185 mm respectively; there were significant differences among the four manufacturers for ASSD, SMCD and visual scores (p < 0.05). And the time for bilateral automatic MC localization was 8.52s (± 0.97s).
In this study, a CNN model was developed for automatic MC localization, and external testing of large sample on multicenter CBCT images showed its excellent clinical application potential.
开发并验证一种基于卷积神经网络（CNN）的深度学习（DL）模型，用于在多中心锥形束计算机断层扫描（CBCT）图像上定位下颌管（MC）。
在本研究中，收集了来自多个中心的总共1056例CBCT扫描。其中，使用一个制造商的836例CBCT扫描来开发CNN模型（训练集：验证集：内部测试集 = 640:360:36），并对来自其他四个制造商的220例CBCT扫描的外部测试数据集进行测试。卷积模块使用Conv + InstanceNorm + LeakyReLU堆栈构建。平均对称表面距离（ASSD）和对称平均曲线距离（SMCD）用于对该模型的内部测试数据和部分外部测试数据进行定量评估。对所有外部测试数据进行视觉评分（1 - 5分），以评估MC定位的准确性和可推广性。比较外部测试数据集在四个制造商之间的ASSD、SMCD和视觉评分的差异。记录手动和自动MC定位的时间。
对于内部测试数据集，平均ASSD和SMCD分别为0.486毫米和0.298毫米。对于外部测试数据集，86.8%的CBCT扫描视觉评分≥4分；40例视觉评分≥4分的CBCT扫描的平均ASSD和SMCD分别为0.438毫米和0.185毫米；四个制造商之间的ASSD、SMCD和视觉评分存在显著差异（p < 0.05）。双侧自动MC定位的时间为8.52秒（±0.97秒）。
在本研究中，开发了一种用于自动MC定位的CNN模型，并且在多中心CBCT图像上的大样本外部测试显示了其优异的临床应用潜力。
