Desarrollo de un aplicativo para recolección de datos para diagnóstico automático de desórdenes mentales (siguiendo la clasificación CIE-10).

Abstract

This paper aims to the development of a web application using the UML-based web engineering methodology (UWE) and the application of a machine learning algorithm kNN to assist professionals in the area of Mental health, for the diagnosis of mental disorders. This tool pretends to facilitate the diagnosis of a patient, and overcome the limitations mainly caused for the extensive variety of mental disorders, according to the ICD-10; the ambiguity of the relationships between symptoms and diseases, the lack of registers of real cases as a reference for new cases, etc. The web application will be used as a platform for the digitalization and acquisition of data from patients that suffer some type of mental disorder, these data will be used for training a first Machine-learning model, whose purpose will be to get the automatic diagnosis of new cases, through the symptomatology of the patient. The algorithm reached 44% success, due to the low cardinality and the dispersion of the characteristics of each instance.