NCT05439447

A Study on Building a Voice Cohort for the Development of a Non-face-to-face Machine Learning Diagnostic and Monitoring Platform Using Voice Analysis and Various Sensors in Patients With Dysphagia.

• Status: Completed
• Phase: —
• Study type: Observational
• Enrollment: 259 (actual)
• Sponsor: Seoul National University Hospital · Academic / Other
• Sex: All
• Age: 19 Years
• Healthy volunteers: Accepted

Summary

Collection of basic data for real-time analysis and monitoring by measuring what changes in voice occur according to the state of dysphagia using voice analysis and various sensors.

Detailed description

* Design : Prospective cohort study * Setting : Hospital rehabilitation department * Inclusion criteria of the patient group : Patients scheduled for VFSS examination * Intervention * This study will be conducted after collecting basic information such as age, gender, and disease of patients who met the inclusion criteria among patients scheduled to undergo a video fluoroscopic swallowing test (VFSS test) for the diagnosis of dysphagia, and after obtaining guidance and consent through the study description will do. * Before the subject's VFSS test, "Ah for 5 seconds", after clearing the throat, "Ah for 5 seconds", and briefly cut with a slightly high sound, "Ah. oh Ah.", shut lips lightly and read "Ummm\~\~\~\~", read a short sentence. * While the subject makes the above sound, voice is recorded, video is taken, and integrated sensors such as a microphone (sound detection) sensor is attached to measure changes in each sensor. * After the VFSS test, participants will repeat the above sound and record the same twice. (As before the test, imaging of the patient's swallowing is also performed.) * Using voice analysis, various sensor values, and image information, a voice indicator that can predict the amount of suction and residue on the VFSS test is found.

Conditions

• Deglutition Disorders

Interventions

Type	Name	Description
DIAGNOSTIC_TEST	Voice recording and sensor-based data collection	Pre- and post-swallowing voice recordings (sustained phonation "ah" for 5 seconds, short phonation, humming, and sentence reading) were collected using a microphone (AKG C 411L) attached to suprahyoid muscles. For VFSS patients, recordings were performed before and after VFSS examination. For healthy controls, recordings were performed before and after yogurt intake. No therapeutic intervention was performed.

Timeline

• Start date: 2023-04-07
• Primary completion: 2024-12-18
• Completion: 2024-12-18
• First posted: 2022-06-30
• Last updated: 2026-02-25

Locations

1 site across 1 country: South Korea