基于深度学习的咬翼片自动图表归档:多中心研究提升临床诊断
本研究提出深度学习方法对咬翼片进行自动图表归档,使用德国与荷兰1045张咬翼片训练验证、斯洛伐克216张外部测试。分层Mask DINO在种植体、冠、桥体、充填、根管治疗、龋损和牙石分类中F1分数为0.944至0.749,特异度均高于0.95,显示提升图表归档效率与准确性的潜力。
要点
- 分层Mask DINO对种植体、冠、桥体、充填F1分数均高于0.91。
- 根管治疗F1分数最高为0.988,龋损和牙石较低为0.749和0.758。
- 各发现分类特异度均高于0.95,可减少手工标注工作量。
核心结论
- 基于深度学习的咬翼片综合评估方法有潜力提升牙科实践中图表归档的效率和准确性。
研究设计与样本
- 多中心研究,使用德国和荷兰的1045张咬翼片进行训练和验证,斯洛伐克的216张进行外部测试。
- 由两名牙医、一名博士研究员和一名龋病专家进行标注。
- 开发分层Mask DINO用于多类分层端到端实例分割,并与未修改的Mask DINO、SparseInst和Mask R-CNN比较。
- 模型性能使用F1分数、敏感度、特异度、精确率、平均精度均值(mAP)和受试者工作特征曲线下面积(AUC)评估。
主要发现
- Mask DINO模型在牙齿分割和标记中精确率、敏感度和F1分数均高于0.96。
- 分层Mask DINO在牙科发现分类中优于其他模型。
- 种植体、冠、桥体、充填、根管治疗、龋损和牙石的F1分数分别为0.944、0.918、0.952、0.956、0.988、0.749和0.758,特异度均高于0.95。
关键数值与推荐
- 分层Mask DINO对种植体、冠、桥体、充填、根管治疗、龋损和牙石的F1分数分别为0.944、0.918、0.952、0.956、0.988、0.749和0.758。
- 上述所有牙科发现分类的特异度均高于0.95。
- Mask DINO模型牙齿分割和标记的精确率、敏感度和F1分数均高于0.96。
临床怎么用
- 原文未据此给出推荐。
局限性
- 原文未报告局限性。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
OBJECTIVES: Bitewings are commonly used radiographs for visualizing teeth and various dental conditions. Manual labeling and diagnosis on bitewings for chart filing are time-consuming and prone to observer-dependent variations. This multi-center study proposes a deep learning (DL) approach to automate comprehensive chart filing of bitewings. METHODS: A total of 1045 bitewings from Germany and The Netherlands were used for training and validation, and 216 from Slovakia for external testing. Annotations were performed by two dentists, one PhD researcher, and one caries expert. Hierarchical Mask DINO was developed for multi-class hierarchical end-to-end instance segmentation. Unmodified Mask DINO, SparseInst, and Mask R-CNN were used for comparison. Model performance was evaluated using F1-score, sensitivity, specificity, precision, mean average precision (mAP), and area under receiver operating characteristic curve (AUC). RESULTS: Mask DINO models exhibited high effectiveness for tooth segmentation and labeling, achieving precision, sensitivity, and F1-scores above 0.96. Hierarchical Mask DINO outperformed the other models in dental finding classification. F1-scores for implant, crown, pontic, filling, root canal treatment (RCT), caries lesion, and calculus deposit were 0.944, 0.918, 0.952, 0.956, 0.988, 0.749, and 0.758, respectively, with specificities all above 0.95. CONCLUSIONS: This study presented a DL-based approach for comprehensive assessment and diagnosis of bitewings, underlining its potential to enhance the efficiency and accuracy of chart filing in dental practice. CLINICAL SIGNIFICANCE: The proposed model provided fully automated tooth segmentation and numbering, along with comprehensive segmentation of dental conditions. Dental professionals can benefit from this model for reducing manual workload and enhancing clinical diagnosis.
引用
- Cao L, van Nistelrooij N, Chaves ET, et al. Automated chart filing on bitewings using deep learning: enhancing clinical diagnosis in a multi-center study. J Dent. 2025. PMID: 40550354. doi:10.1016/j.jdent.2025.105919
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。