Duration: May 13th, 2020 – July 22nd, 2020 (10 weeks)
Department: Department of Computer and Electrical Engineering and Department of Communicative Disorders
Faculty Mentors: Dr. Gan and Dr. Gosa
Project: Image Analysis to Identify Features from Videofluroscopy during Pediatric Dysphagia
Image Analysis to Identify Features from Videofluroscopy during Pediatric Dysphagia
Background: When infants present with swallowing difficulties, referred to as dysphagia, it is necessary to determine if airway compromise is a component of their feeding difficulties. Assessment for dysphagia can be accomplished using imaging instrumentation technology including: videofluroscopic swallow study (VFSS), fiberoptic endoscopic evaluation, ultrasonography, manometry, scintigraphy, and cervical auscultation. The most common instrumental assessment for dynamic assessment of oropharyngeal swallowing function in pediatric patients is VFSS. VFSS uses x-ray imaging to visualize what is occurring internally during swallowing events. Clinicians look for abnormalities in the oral, pharyngeal, and esophageal phases of swallowing function during recorded events, commenting on: bolus extraction, formation, and propulsion, spillover prior to the swallow, oral residue, oral transit time, timing of pharyngeal swallow initiation, strength of pharyngeal swallow, pharyngeal residue, presence of laryngeal penetration, aspiration, or nasopharyngeal backflow, pharyngeal transit time, opening of cricopharyngeal sphincter, clearance of bolus through cervical esophagus, retrograde movement through cervical esophagus to pyriform sinuses. Currently, to identify these features, clinicians must participate in frame-by-frame analysis which is labor intensive, time consuming, and not-realistic in a clinical setting where multiple VFSS are completed each day. However, there are currently no available imaging processing tools to automate these tasks for clinicians, providing the motivation for this project (University of Alabama REU).
REU Participants Role: Students will be trained on the swalling and clinical feautures observed during VFSS and image processing techniques. Using existing VFSS video data, the students will investigate methods to extract clinically relevant features from imaging data, applying machine learning techniques towards the goal of automated and reliable identification and detection of airway compromise characteristics.