RESUMO
Objective@#To research the effectiveness of deep learning techniques in intelligently diagnosing dental caries and periapical periodontitis and to explore the preliminary application value of deep learning in the diagnosis of oral diseases@*Methods@#A dataset containing 2 298 periapical films, including healthy teeth, dental caries, and periapical periodontitis, was used for the study. The dataset was randomly divided into 1 573 training images, 233 validation images, and 492 test images. By comparing various neural network models, the MobileNetV3 network model with better performance was selected for dental disease diagnosis, and the model was optimized by tuning the network hyperparameters. The accuracy, precision, recall, and F1 score were used to evaluate the model's ability to recognize dental caries and periapical periodontitis. Class activation map was used to visualization analyze the performance of the network model@*Results@#The algorithm achieved a relatively ideal intelligent diagnostic effect with precision, recall, and accuracy of 99.42%, 99.73%, and 99.60%, respectively, and the F1 score was 99.57% for classifying healthy teeth, dental caries, and periapical periodontitis. The visualization of the class activation maps also showed that the network model can accurately extract features of dental diseases.@*Conclusion@#The tooth lesion detection algorithm based on the MobileNetV3 network model can eliminate interference from image quality and human factors and has high diagnostic accuracy, which can meet the needs of dental medicine teaching and clinical applications.
RESUMO
@#Dental caries detector is a kind of diagnostic tool specifically designed for dental professionals to detect and monitor the early occurrence and development of dental caries. They are widely used in the clinic because of their advantages of rapid detection, flexible applications, ease of carrying, intuitive detection results and lack of pain for the patient. However, due to the different types and principles of the instruments produced by various instrument manufacturers, the clinical application range, sensitivity and specificity of test results also show significant differences. In terms of the current clinical application effects, although the DIAGNOdent caries detector has the widest range of clinical use, the accuracy of its detection results needs to be improved because it is affected by factors such as pigments and dental materials. The Canary System caries detector can effectively avoid the interference of the above factors, but its classification of the degree of caries is not clear. The DIAGNOcam caries detector can effectively detect early caries, but it has low reliability for occlusal caries detection. The existing dental caries detectors on the market can be used only as clinical auxiliary tools, and the accuracy of the detection results and comprehensiveness of the detection range need further improvement. With the application of the new multispectral near-infrared scanning fiber endoscope (NIR-SFE) and high-frequency ultrasound imaging (HFUS) in the detection of dental caries, a more efficient and accurate diagnosis of dental caries is possible in the future. To this end, we still need to continue exploring new technology to help clinicians complete the early diagnosis and treatment of dental caries to improve the quality of life of their patients.