Diagnosis of mTBI in an ER Setting Using Eye-Tracking and Virtual Reality Technology: An Exploratory Study

TL;DR

This study explored using eye-tracking and VR technology to diagnose mild traumatic brain injury in emergency rooms but found it ineffective.

Contribution

The study is the first to explore ET/VR technology for mTBI diagnosis in ERs with a systematic accuracy assessment.

Findings

ET/VR glasses were mostly unproblematic to use in the ER setting.

No significant differences in oculomotor function were found between mTBI and non-TBI patients.

VR goggles did not correlate with mTBI detection based on recorded parameters.

Abstract

Background: The aim of this study was to systematically explore point-of-care biomarkers as diagnostic indicators for the detection and exclusion of mild traumatic brain injury (mTBI) in an emergency room (ER) setting using Eye-Tracking and Virtual Reality (ET/VR) technology. The primary target group included patients who had suffered an acute trauma to the head and presented within 24 h to the emergency department. Methods: The BG Unfallkrankenhaus Berlin and the BG Klinikum Hamburg participated in this explorative, prospective, single-arm accuracy study. This study included patients who presented to the emergency department with suspected mTBI and were examined using ET/VR glasses. All further steps corresponded to clinical routine (e.g., decision on hospital admission, imaging diagnostics). After the completion of treatment, the patients were divided into mTBI and non-TBI subgroups by consensus between two independent clinical experts, who were blinded to the results of the index test (examination using ET/VR glasses) in the form of a clinical synopsis. The diagnosis was based on all clinical, neurological, neurofunctional, neuropsychological, and imaging findings. Routine trauma and neurological history, examination, and diagnosis were performed in each case. All statistical analyses were performed with exploratory intent. Results: The use of ET/VR glasses was found to be predominantly unproblematic. Two of the fifty-two analyzed parameters can be statistically distinguished from a random decision. No difference in oculomotor function was found between the two subgroups, and no correlations between the parameters recorded by the VR goggles and the detection of mTBI were found. Conclusions: At present, the use of VR goggles for the diagnosis of mTBI in an ER setting cannot be recommended.

Figures