Deep learning based prediction of extraction difficulty for mandibular third molars

Abstract

This paper proposes a convolutional neural network (CNN)-based deep learning model for predicting the difficulty of extracting a mandibular third molar using a panoramic radiographic image. The applied dataset includes a total of 1053 mandibular third molars from 600 preoperative panoramic radiographic images. The extraction difficulty was evaluated based on the consensus of three human observers using the Pederson difficulty score (PDS). The classification model used a ResNet-34 pretrained on the ImageNet dataset. The correlation between the PDS values determined by the proposed model and those measured by the experts was calculated. The prediction accuracies for C1 (depth), C2 (ramal relationship), and C3 (angulation) were 78.91%, 82.03%, and 90.23%, respectively. The results confirm that the proposed CNN-based deep learning model could be used to predict the difficulty of extracting a mandibular third molar using a panoramic radiographic image.

Conflict of interest statement

The authors declare no competing interests.

Figures

Figure 1

Predicted Pederson difficulty score (PDS) distribution of the actual PDS. The distribution of predicted PDS was generally close to the actual PDS. Whereas it performed well for PDS 4–7, it overestimated the cases of PDS 3 and underestimated the cases of PDS 8 and 9.

Figure 2

Confusion matrix showing the classification results for each criterion.

Figure 3

Example probability distribution inferred by the proposed model. The blue-dashed line indicates the actual score, and the red-dashed line indicates the expectation of the predicted PDS. If the model obtains a probability distribution as depicted in (A), it will be received a score of 1. Conversely, if a probability distribution similar to that depicted in (B) is obtained, its score will be miscalculated as 2.

Figure 4

Preprocessing of panoramic images.

Figure 5

Entire diagnosis process adopted in this study.