Integrating Digital Twin Technology into Smart Dental Healthcare: A Framework and Applications

This paper presents a new digital twin framework for simulating the real dental problem and many diseases to save the teeth pervious damage. A new digital twin framework is developed based on two ways of supervised data stream that begins when question sensors give related information to the framework processor and proceed. A digital twin framework differs the traditional classification methods using machine learning or deep learning in classification the disease damage type, degree level, and it also can predict the probable problems and suggest solution for saving the teeth life. It also simulation is the scale of simulation size with examining the teeth behavior and evolution. The experiment is applied on Dental panoramic x-rays that use a very small dose of ionizing radiation to capture the entire mouth in a single image. They are typically performed by dentists and oral surgeons, and can be used to plan treatment for dentures, braces, extractions and implants. The dataset includes X-rays images that is used for improve the accuracy classification. The proposed solution model enlarges dataset for improving the accuracy of dental X-rays images with respect four augmentation models to achieve the dataset size to 29.696 X-rays images for each class and the cumulative total dataset includes 89.088 X-rays images. The accuracy result reaches VGG16 achieves to 95.6%, GoogleNet achieves to 97.2%, and AlexNet achieves to 99.98%.