Uso de algoritmos de machine learning para auxiliar no diagnóstico de pacientes portadores de disfonia espasmódica

Abstract

Spasmodic dysphonia (SD) is a rare voice disorder of neurological etiology that typically affects patients around the age of 30, impairing the process of vocal production and potentially being exacerbated by stressful situations. Given its neurological origin, diagnosis by speech-language pathology professionals is challenging and highly dependent on the evaluator’s expertise in recognizing the disorder. This may prevent patients from receiving optimal treatment, leading to the persistence of symptoms. With technological advances, the search for non-invasive methods for identifying voice disorders can be advantageous in supporting the diagnosis of SD. This study aims to evaluate the use of machine learning techniques to classify voice signals between patients with SD and healthy individuals.