Dental X-Ray Image Analysis for Diagnosis Utilized Convolutional Neural Network Using Real-Time Performance

The diagnostic methods available to dentists for caries detection have multiplied in dental X-ray images. In the research, a classification system was designed using Convolutional Neural Network (CNN) analysis to obtain the best parameters for the system. CNNs can be trained to detect and classify dental conditions, such as cavities which can assist dentists in diagnosing and treatment planning. The large number of X-ray images allows dentists to make quicker decisions, reducing the time and risk of interpreting the X-rays. After testing three different image sizes, it was found that an image size of 224 × 224 pixels yielded the best testing results and graphs. This is because each pixel contains information about the image, and larger image sizes provide more detailed information. Therefore, over time, monitoring with CNNs can help dentists track the progress of ongoing treatments and interventions. To evaluate the effectiveness of these models, experiments were conducted on a manually annotated dataset known as Dental X-ray Images. The experimental results demonstrate that the CNN achieved an accuracy rate of 98.60%, surpassing other existing models such as ResNet, U-Net, and DenseNet.