Towards dental diagnostic systems: Synergizing wavelet transform with generative adversarial networks for enhanced...

Area of Science:

• Pediatric Dentistry
• Medical Imaging Analysis
• Artificial Intelligence in Healthcare

Background:

• Early detection of dental diseases is crucial for pediatric oral health.
• Current diagnostic systems require enhancement for improved accuracy and efficiency.
• Precision diagnostics are transforming pediatric dental care.

Purpose of the Study:

• To develop an innovative dental diagnostic system using image data fusion (IDF).
• To enhance the accuracy of detecting dental diseases in pediatric patients.
• To explore the integration of AI-powered diagnostics into dental X-ray scanners.

Main Methods:

• Synergistic integration of Discrete Wavelet Transform (DWT) and Generative Adversarial Networks (GANs) within an Image Data Fusion (IDF) framework.
• Utilizing DWT for image decomposition to highlight pathological features and GANs for data augmentation.
• Employing an Artificial Neural Network (ANN) for classifying dental diseases from enhanced radiographic images.

Main Results:

• Achieved an accuracy rate of 0.897, with 0.905 precision, 0.897 recall, and 0.968 specificity.
• Demonstrated significant improvement in the informativeness of dental panoramic radiographs.
• Validated the robustness of the ANN in classifying dental diseases.

Conclusions:

• The DWT-GANs-IDF framework offers a highly accurate method for pediatric dental disease diagnosis.
• Integration into dental X-ray scanners via lightweight models and cloud solutions is feasible.
• This system has the potential to revolutionize dental care by enabling real-time detection and improving treatment outcomes.