A Pilot Study on Video-based Eye Movement Assessment of the NeuroEye Examination

Abstract

The ability to perform automated neurological assessment could enhance diagnosis and treatment in the pre-hospital setting such as during telemedicine or emergency medical services (EMS) encounters. In this pilot study we present the preliminary analysis of a video-based eye movement assessment workflow (NeuroGaze) for neurological eye examinations. Our workflow uses a pupil detector to detect the center of the pupil for both eyes for a given video and determine the conjugacy of eye movement. The study evaluated the proposed workflow on 18 healthy volunteers while performing a digitally adapted neurological eye examination (NeuroEye) to assess conjugacy of gaze in all cardinal positions. The NeuroGaze performance was compared to the COTS eye tracker. The video-based eye movement assessment workflow reported an average Pearson correlation coefficient of 0.84, compared to the COTS eye tracker with an average of 0.95. The preliminary analysis indicates that NeuroGaze can be used to determine the individuals ability to perform conjugate eye movements. Future study will assess the performance of NeuroGaze in patients with pathological eye movements.