Diseases Classification Utilizing Tooth X-ray Images Based On Convolutional Neural Network

Artificial Intelligence (AI), Deep Learning (DL), and Convolutional Neural Network (CNN) related technologies have seen widespread applications in various fields, such as in finance, military and medicine. In the medical field in particular, by utilizing the technology of CNN, there are many potential use cases in the real world, such as lung cancer detection, malignant tumor classification and tooth decay diagnosis. In this paper, we proposed a type of CNN architecture that used DL to classify four categories of tooth X-rays images: normal teeth, implants, fillings and abnormal teeth (cavities). The requirements for neural network architecture is also very important in this research, so we changed the CNN architecture and the parameters of the test set many times to achieve optimal performance. The preliminary results showed the highest detection accuracy rate of the four categories were normal teeth at 87%, implants and fillings at 98% and cavities at 89%. We were able to achieve an average accuracy at 93.04%. Thus we believed that this result could apply in periodontology dentistry field in the near future.