Main Article Content
The aim of this paper is to develop a methodology, through studies on Computer Vision techniques, for the automatic identification of dental cysts in panoramic radiography images, providing Dental professionals with an alternative aid in the interpretation of these images. In addition segmentation techniques are applied in the inner region of the jaws, seeking to separate the regions with a greater possibility of cyst. The objective of this work is to design an intelligent system that supports the diagnosis of Dental Cyst using convolutional neural networks in order to help detect Dental Cyst at an early stage. The research method applied in this study consists of model design, where built and trained two convolutional neural network architectures, supporting 80% of the dataset with a total of 775 images with four image categories, and proposal validation, where we work with the remaining 20% of the dataset. Our results show that the ResNet50 architecture achieved the best classification with an accuracy of 98%.