Desarrollo de un sistema de análisis de imágenes de rayos x para la detección de afecciones dentales utilizando visión artificial

Abstract

This project presents the development of an automated system for the analysis of dental X-ray images using convolutional neural networks (CNN) for the detection of dental conditions, specifically caries and periapical lesions. The proposal is aimed at improving the accuracy of dental diagnoses through the use of advanced computer vision techniques. The system is based on a dataset of annotated dental X-rays and employs a supervised machine learning approach to train the CNN. PyTorch was used as the development framework, Jupyter Notebook for writing and running the code, and Anaconda for environment and dependency management. Throughout the development, the Faster R-CNN object detection model was evaluated, recognized for its high accuracy in diagnosing dental conditions. Additionally, Data Augmentation techniques were applied to enhance the diversity of the dataset and increase the robustness of the model. The work includes image preprocessing techniques and manual data annotation using specialized tools like LabelMe. Finally, the use of this system is proposed in the clinical field, where it could assist dentists in the quick and accurate identification of conditions, optimizing diagnosis time and reducing dependence on the specialist's visual experience.