An Explainable Deep Learning Model to Prediction Dental Caries Using Panoramic Radiograph Images

Abstract

Dental caries is the most frequent dental health issue in the general population. Dental caries can result in extreme pain or infections, lowering people's quality of life. Applying machine learning models to automatically identify dental caries can lead to earlier treatment. However, physicians frequently find the model results unsatisfactory due to a lack of explainability. Our study attempts to address this issue with an explainable deep learning model for detecting dental caries. We tested three prominent pre-trained models, EfficientNet-B0, DenseNet-121, and ResNet-50, to determine which is best for the caries detection task. These models take panoramic images as the input, producing a caries-non-caries classification result and a heat map, which visualizes areas of interest on the tooth. The model performance was evaluated using whole panoramic images of 562 subjects. All three models produced remarkably similar results. However, the ResNet-50 model exhibited a slightly better performance when compared to EfficientNet-B0 and DenseNet-121. This model obtained an accuracy of 92.00%, a sensitivity of 87.33%, and an F1-score of 91.61%. Visual inspection showed us that the heat maps were also located in the areas with caries. The proposed explainable deep learning model diagnosed dental caries with high accuracy and reliability. The heat maps help to explain the classification results by indicating a region of suspected caries on the teeth. Dentists could use these heat maps to validate the classification results and reduce misclassification.

Keywords: Grad-CAM; caries; deep learning; dental health; explainable deep models.

Figure 1

A block representation of the material and method used in the study.

Figure 2

Demonstration of tooth samples used in our study: (a) Caries teeth and (b) non-caries teeth.

Figure 3

A block representation of the EfficientNet-B0 deep learning model.

Figure 4

A block diagram representation of the ResNet model depicting its working structure.

Figure 5

A block diagram representation of the DenseNet deep learning model.

Figure 6

Model performance curves during training: (a) Training accuracy values for all models, (b) validation accuracy values for all models.

Figure 7

Confusion matrices and ROC curves obtained by the models on the test images: (a) EfficientNet-B0, (b) DenseNet-121, and (c) ResNet-50.

Figure 8

A block diagram representing the Grad-CAM algorithm’s working structure.

Figure 9

Heat maps of various test images were obtained using the Grad-CAM method. Red lines indicate the expert-marked caries regions.

Figure 10

Original images and heatmaps of some test images misclassified by the proposed model.

Figure 11

An illustration of the image that the model misclassified due to artifacts.

Figure 12

The proposed model misclassified the image of Sample 5. The red line and arrows indicate the early-stage caries area, as marked by the expert.

Figure 13

A clinical use case of the proposed caries detection model to automatically detect caries areas on the entire panoramic tooth image.