A Novel Kinematics-based Approach to Inform Vestibular Rehabilitation

Sarah Li

Mentor: Dr. Kathleen Cullen

Supervisor: Omid Zobeiri

Department of Biomedical Engineering, Johns Hopkins University

Nearly 100 million American adults have experienced vestibular dysfunction, which

causes patients to experience position, balance, and movement issues. The current standard

treatment for vestibular disorders is traditional vestibular rehabilitation therapy (VRT).

However, traditional VRT relies on limited observations in the clinic. Accordingly, there exists a

significant need for objective quantification of patient performance of VRT exercises. This study

seeks to evaluate the value of kinematics-based analysis in VRT to improve the quality,

management, and delivery of VRT. Shimmer3 inertial sensors were used to record kinematic

data of vestibular schwannoma surgery patients and age-sex matched healthy controls during

VRT exercises. Patients were tested pre-operatively and during six postoperative clinic visits.

Data from gaze stabilization exercises were compared between patients and controls. We found

that one week post-surgery, patients moved their heads significantly slower than healthy controls

did during execution of gaze exercises. Following the second postoperative assessment, patients

moved their heads quicker over time at a velocity similar to the healthy controls. One week

following vestibular schwannoma surgery, patient performance of gaze exercises was poor in

comparison to healthy controls, but following the second week after surgery patient performance

improved. A mobile app was also developed to allow users to access objective feedback about

performance of VRT exercises. The app offered feedback to users based on how much

kinematics-based indicators deviated from a specific value. These findings allow for future

leveraging of technology to better personalize VRT programs to fit patient needs and ultimately

optimize treatment outcomes for vestibular patients.