Automatic Voice Analysis for Dysphagia Screening in Neurological Patients

The proposed study suggests using automatic voice analysis and machine learning algorithms to develop a dysphagia screening tool for neurological patients. The research involves patients with Parkinson's disease, stroke, and amyotrophic lateral sclerosis, both with and without dysphagia, along with healthy individuals. Participants perform various vocal tasks during a single recording session. Voice signals are analysed and used as input for machine learning classification algorithms. The significance of this study is that oropharyngeal dysphagia, a condition involving swallowing difficulties in the transit of food or liquids from the mouth to the esophagus, generates malnutrition, dehydration, and pneumonia, significantly contributing to management costs and hospitalization durations. Currently, there is a lack of rapid and effective dysphagia screening methods for healthcare personnel, with only expensive invasive tests and clinical scales in use.

NCT Number	NCT06219200
Lead Sponsor	Istituti Clinici Scientifici Maugeri SpA Collaborators: Politecnico di Milano
Conditions	Deglutition Disorders, Neurological Disorder
Enrollment	400 participants
Start Date	2023-10-11
Primary Completion	2025-12 (estimated)
Study Completion	2025-12 (estimated)
Updated on ClinicalTrials.gov	2025-02-20