在全景片上检测分离的根管器械：LSTM 和 CNN 深度学习方法的比较。
Detection of the separated root canal instrument on panoramic radiograph: a comparison of LSTM and CNN deep learning methods.
机构信息
Department of Dentomaxillofacial Radiology, Istanbul Okan University, Faculty of Dentistry, İstanbul, Turkey.
Department of Endodontics, Bahcesehir University, Faculty of Dentistry, İstanbul, Turkey.
出版信息
Dentomaxillofac Radiol. 2023 Feb;52(3):20220209. doi: 10.1259/dmfr.20220209. Epub 2023 Jan 25.
OBJECTIVES
A separated endodontic instrument is one of the challenging complications of root canal treatment. The purpose of this study was to compare two deep learning methods that are convolutional neural network (CNN) and long short-term memory (LSTM) to detect the separated endodontic instruments on dental radiographs.
METHODS
Panoramic radiographs from the hospital archive were retrospectively evaluated by two dentists. A total of 915 teeth, of which 417 are labeled as "separated instrument" and 498 are labeled as "healthy root canal treatment" were included. A total of six deep learning models, four of which are some varieties of CNN (Raw-CNN, Augmented-CNN, Gabor filtered-CNN, Gabor-filtered-augmented-CNN) and two of which are some varieties of LSTM model (Raw-LSTM, Augmented-LSTM) were trained based on several feature extraction methods with an applied or not applied an augmentation procedure. The diagnostic performances of the models were compared in terms of accuracy, sensitivity, specificity, positive- and negative-predictive value using 10-fold cross-validation. A McNemar's tests was employed to figure out if there is a statistically significant difference between performances of the models. Receiver operating characteristic (ROC) curves were developed to assess the quality of the performance of the most promising model (Gabor filtered-CNN model) by exploring different cut-off levels in the last decision layer of the model.
RESULTS
The Gabor filtered-CNN model showed the highest accuracy (84.37 ± 2.79), sensitivity (81.26 ± 4.79), positive-predictive value (84.16 ± 3.35) and negative-predictive value (84.62 ± 4.56 with a confidence interval of 80.6 ± 0.0076. McNemar's tests yielded that the performance of the Gabor filtered-CNN model significantly different from both LSTM models ( < 0.01).
CONCLUSIONS
Both CNN and LSTM models were achieved a high predictive performance on to distinguish separated endodontic instruments in radiographs. The Gabor filtered-CNN model without data augmentation gave the best predictive performance.
目的
根管治疗中分离的根管器械是一项具有挑战性的并发症。本研究旨在比较两种深度学习方法，即卷积神经网络（CNN）和长短时记忆（LSTM），以检测牙片上分离的根管器械。
方法
两位牙医回顾性评估了医院档案中的全景片。共纳入 915 颗牙齿，其中 417 颗标记为“分离器械”，498 颗标记为“健康根管治疗”。基于几种特征提取方法，共训练了 6 种深度学习模型，其中 4 种为 CNN 的不同变体（Raw-CNN、Augmented-CNN、Gabor 滤波-CNN、Gabor-filtered-augmented-CNN），2 种为 LSTM 模型的不同变体（Raw-LSTM、Augmented-LSTM），并应用或不应用增强程序。通过 10 折交叉验证，比较了模型在准确性、敏感性、特异性、阳性预测值和阴性预测值方面的诊断性能。采用 McNemar 检验来确定模型性能之间是否存在统计学显著差异。绘制受试者工作特征（ROC）曲线，通过探索模型最后决策层的不同截断水平，评估最有前途的模型（Gabor 滤波-CNN 模型）的性能质量。
结果
Gabor 滤波-CNN 模型表现出最高的准确性（84.37±2.79）、敏感性（81.26±4.79）、阳性预测值（84.16±3.35）和阴性预测值（84.62±4.56，置信区间为 80.6±0.0076）。McNemar 检验表明，Gabor 滤波-CNN 模型的性能与两种 LSTM 模型均有显著差异（<0.01）。
结论
CNN 和 LSTM 模型在区分牙片上分离的根管器械方面均具有较高的预测性能。未经数据增强的 Gabor 滤波-CNN 模型具有最佳的预测性能。
Zotero 插件