Department of Endodontics, Dental School, Hamadan University of Medical Sciences, Hamadan, Iran.
Department of Endodontics, Dental Research Center, Hamadan University of Medical Sciences, Hamadan, Iran.
BMC Oral Health. 2024 May 17;24(1):574. doi: 10.1186/s12903-024-04235-4.
To develop and validate a deep learning model for automated assessment of endodontic case difficulty from periapical radiographs.
A dataset of 1,386 periapical radiographs was compiled from two clinical sites. Two dentists and two endodontists annotated the radiographs for difficulty using the "simple assessment" criteria from the American Association of Endodontists' case difficulty assessment form in the Endocase application. A classification task labeled cases as "easy" or "hard", while regression predicted overall difficulty scores. Convolutional neural networks (i.e. VGG16, ResNet18, ResNet50, ResNext50, and Inception v2) were used, with a baseline model trained via transfer learning from ImageNet weights. Other models was pre-trained using self-supervised contrastive learning (i.e. BYOL, SimCLR, MoCo, and DINO) on 20,295 unlabeled dental radiographs to learn representation without manual labels. Both models were evaluated using 10-fold cross-validation, with performance compared to seven human examiners (three general dentists and four endodontists) on a hold-out test set.
The baseline VGG16 model attained 87.62% accuracy in classifying difficulty. Self-supervised pretraining did not improve performance. Regression predicted scores with ± 3.21 score error. All models outperformed human raters, with poor inter-examiner reliability.
This pilot study demonstrated the feasibility of automated endodontic difficulty assessment via deep learning models.
开发并验证一种基于根尖片的牙髓病例难度自动评估的深度学习模型。
从两个临床站点收集了 1386 张根尖片数据集。两名牙医和两名牙髓病学家使用 Endocase 应用程序中美国牙髓病协会病例难度评估表的“简单评估”标准对根尖片进行了难度标注。分类任务将病例标记为“简单”或“困难”，而回归则预测总体难度评分。使用卷积神经网络（即 VGG16、ResNet18、ResNet50、ResNext50 和 Inception v2），基线模型通过从 ImageNet 权重进行迁移学习进行训练。其他模型通过在 20295 张未标记的牙科射线照片上使用自我监督对比学习（即 BYOL、SimCLR、MoCo 和 DINO）进行预训练，以在没有手动标签的情况下学习表示。这两种模型都通过 10 折交叉验证进行评估，并在保留测试集上与七名人类检查者（三名普通牙医和四名牙髓病学家）的表现进行比较。
基线 VGG16 模型在分类难度方面的准确率达到 87.62%。自我监督预训练并没有提高性能。回归预测的分数误差为±3.21 分。所有模型的表现均优于人类评分者，且评分者之间的可靠性较差。
这项初步研究表明，通过深度学习模型自动评估牙髓病例难度是可行的。
