DysphagiaScan is an Automatic Audio-based Dysphagia Screening device that hopes to increase early detection, accessibility, and clinical efficiency for patients with swallowing impairments. It uses noninvasive acoustic sensing through a contact microphone attached to the neck and a machine learning model for classification of normal and dysphagic swallows. This enables clinicians, or potentially caregivers in remote settings, to screen for swallowing difficulties quickly and reliably.
This device was created as part of a UCSD Bioengineering Senior Design Project in 2025, in collaboration with clinical experts and data from real patients.
Dysphagia is the medical term for difficulty swallowing. It is especially common in individuals with stroke, Parkinson’s disease, ALS, and aging-related conditions. Missed diagnosis can lead to aspiration pneumonia, malnutrition, and reduced quality of life. While gold-standard methods like FEES and VFSS are effective for diagnosing and screening for dysphagia, they are invasive, expensive, and require specialized expertise. DysphagiaScan aims to address these limitations.
This page was made by Jade Chng