Sistema de auxílio ao diagnóstico de doenças bucais: uma abordagem utilizando visão computacional

Abstract

The decision-making process is a critical point in all professions; for healthcare professionals, the resolution reached will influence the course of treatment and the lives of the people involved. Diagnostic support systems aim to facilitate this process. Systems based on machine learning and computer vision may serve as an effective alternative approach during treatment selection after lesion diagnosis by a healthcare professional. These applications may potentially discover similarities between lesion images that could remain unperceived by the human eye. In addition, morphological alterations, such as mouth lesions, can develop and manifest in different forms and complicate the diagnostic process. Thus, the knowledge and experience of the healthcare professional can be supplemented with decision support systems established by clinical evidence, in order to promote evidence-based practices. In the form of a web-based application, the present work introduces a prototype of a diagnostic support system for the retrieval of mouth lesion images, proposing a means for assisting healthcare professionals.