Pulp calcification identification on cone beam computed tomography: an artificial intelligence pilot study

Area of Science:

• Dentistry
• Radiology
• Artificial Intelligence

Background:

• Pulp calcification detection is crucial for root canal treatment planning.
• Cone beam computed tomography (CBCT) is a common imaging modality in dentistry.
• Automated identification of pulp calcification can enhance diagnostic efficiency.

Purpose of the Study:

• To evaluate the effectiveness of a deep neural network (DNN) for automatic pulp calcification identification.
• To assess the performance of a U-Net based DNN model on CBCT images.

Main Methods:

• Retrospective analysis of 150 CBCT images.
• Manual annotation of pulp calcification by experienced endodontists.
• Development and validation of a U-Net DNN model using cross-validation.
• Performance evaluation using sensitivity, specificity, accuracy, and AUC.

Main Results:

• The DNN model achieved a sensitivity of 75.91%, specificity of 68.88%, accuracy of 72.78%, and AUC of 73.68%.
• The model demonstrated "good" performance in sensitivity, accuracy, and AUC, and "fair" in specificity.
• The results indicate the potential of AI in identifying pulp calcification.

Conclusions:

• The proposed DNN method shows potential for identifying pulp calcification on CBCT images.
• Further research will focus on expanding the dataset and refining the model for improved clinical applicability.
• AI integration is expected to increase diagnostic efficiency and aid in preoperative case difficulty assessment for root canal treatments.