A Machine Learning-Based Voice Analysis for the Detection of Dysphagia Biomarkers

摘要

A Machine-Learning process for selecting optimal biomarkers that identify Dysphagia is presented. The effectiveness of said biomarkers is confirmed by an ensemble of Classifiers that correctly distinguish between Healthy and Dysphagic patients with high Accuracy. An overview of the clinical meaning of the biomarkers found is presented in the Discussion, corroborating and further refining the previous studies in the matter. RASTA Processing for speech and spectral energy distribution are the main domains for detecting Dysphagia in the voice.