Department of Advanced Prosthodontics, Graduate School of Medical and Dental Sciences, Tokyo Medical and Dental University, Tokyo, Japan.
Department of Prosthodontics, Faculty of Dentistry, Mansoura University, Mansoura, Egypt.
J Prosthodont. 2024 Aug;33(7):645-654. doi: 10.1111/jopr.13853. Epub 2024 Apr 3.
The study aimed to compare the performance of four pre-trained convolutional neural networks in recognizing seven distinct prosthodontic scenarios involving the maxilla, as a preliminary step in developing an artificial intelligence (AI)-powered prosthesis design system.
Seven distinct classes, including cleft palate, dentulous maxillectomy, edentulous maxillectomy, reconstructed maxillectomy, completely dentulous, partially edentulous, and completely edentulous, were considered for recognition. Utilizing transfer learning and fine-tuned hyperparameters, four AI models (VGG16, Inception-ResNet-V2, DenseNet-201, and Xception) were employed. The dataset, consisting of 3541 preprocessed intraoral occlusal images, was divided into training, validation, and test sets. Model performance metrics encompassed accuracy, precision, recall, F1 score, area under the receiver operating characteristic curve (AUC), and confusion matrix.
VGG16, Inception-ResNet-V2, DenseNet-201, and Xception demonstrated comparable performance, with maximum test accuracies of 0.92, 0.90, 0.94, and 0.95, respectively. Xception and DenseNet-201 slightly outperformed the other models, particularly compared with InceptionResNet-V2. Precision, recall, and F1 scores exceeded 90% for most classes in Xception and DenseNet-201 and the average AUC values for all models ranged between 0.98 and 1.00.
While DenseNet-201 and Xception demonstrated superior performance, all models consistently achieved diagnostic accuracy exceeding 90%, highlighting their potential in dental image analysis. This AI application could help work assignments based on difficulty levels and enable the development of an automated diagnosis system at patient admission. It also facilitates prosthesis designing by integrating necessary prosthesis morphology, oral function, and treatment difficulty. Furthermore, it tackles dataset size challenges in model optimization, providing valuable insights for future research.
本研究旨在比较四个预先训练的卷积神经网络在识别涉及上颌骨的七种不同修复场景中的性能，作为开发人工智能（AI）驱动的义齿设计系统的初步步骤。
考虑了七种不同的类别，包括腭裂、牙列缺损上颌骨切除术、无牙颌上颌骨切除术、重建上颌骨切除术、完全有牙、部分无牙和完全无牙。利用迁移学习和微调超参数，使用了四个 AI 模型（VGG16、Inception-ResNet-V2、DenseNet-201 和 Xception）。该数据集由 3541 张预处理的口腔咬合图像组成，分为训练集、验证集和测试集。模型性能指标包括准确性、精度、召回率、F1 分数、受试者工作特征曲线下的面积（AUC）和混淆矩阵。
VGG16、Inception-ResNet-V2、DenseNet-201 和 Xception 的表现相当，最大测试准确率分别为 0.92、0.90、0.94 和 0.95。Xception 和 DenseNet-201 略微优于其他模型，特别是与 Inception-ResNet-V2 相比。Xception 和 DenseNet-201 中大多数类别的精度、召回率和 F1 得分均超过 90%，所有模型的平均 AUC 值在 0.98 到 1.00 之间。
尽管 DenseNet-201 和 Xception 表现出了优异的性能，但所有模型的诊断准确性均超过 90%，这表明它们在牙科图像分析方面具有潜力。该 AI 应用可以帮助根据难度级别分配工作，并在患者入院时开发自动诊断系统。它还通过整合必要的义齿形态、口腔功能和治疗难度，为义齿设计提供便利。此外，它还解决了模型优化中的数据集大小挑战，为未来的研究提供了有价值的见解。
