DEEP LEARNING ALGORITHMS SHOW SOME POTENTIAL AS AN ADJUNCTIVE TOOL IN CARIES DIAGNOSIS

Area of Science:

• Dentistry
• Artificial Intelligence
• Medical Imaging

Background:

• Dental caries detection remains a critical diagnostic challenge.
• Traditional methods for caries detection have limitations in accuracy and efficiency.
• Deep learning (DL) offers potential for automated and improved caries detection.

Purpose of the Study:

• To systematically review and analyze the current literature on the application of deep learning algorithms for dental caries detection.
• To evaluate the performance and diagnostic accuracy of various DL models in identifying dental caries from radiographic images.
• To identify trends, challenges, and future directions in the field of DL for caries detection.

Main Methods:

• A systematic literature search was conducted across major scientific databases (e.g., PubMed, Scopus, Web of Science).
• Studies employing deep learning techniques for the detection of dental caries from dental radiographs (e.g., bitewings, periapical radiographs) were included.
• Data extraction focused on study design, DL model architecture, dataset characteristics, performance metrics (sensitivity, specificity, AUC), and reported outcomes.

Main Results:

• The review identified a growing number of studies utilizing deep learning for caries detection, demonstrating promising results.
• Various convolutional neural network (CNN) architectures were commonly employed, achieving high diagnostic accuracy in several studies.
• Performance varied depending on image quality, dataset size, and specific DL model implementation, highlighting the need for standardization.

Conclusions:

• Deep learning models show significant potential to enhance the accuracy and efficiency of dental caries detection.
• Further research with larger, diverse datasets and standardized evaluation protocols is necessary to validate these findings.
• The integration of DL into clinical practice could revolutionize caries diagnosis and management.