Interpretación automatizada de radiografías de tórax: Un prototipo con Machine Learning orientado desde la Ingeniería de Software

Abstract

This project was built with the intention of carrying out the use of artificial intelligence in projects applied in business environments, describing the various development processes within the development of web applications. From the training of 2 deep learning models for the detection of pneumonia in frontal chest radiographs and with the definition of a software architecture for the desired prototype, the various components of the prototype as a final application are described, designed and implemented. Following certain business rules, the prototype integrates a machine learning model for the detection of pneumonia in frontal chest X-rays, from a web interface, connections to a database and other common practices in this type of web applications.