Abstract
Purpose :
Virtual Reality (VR) headsets are becoming attractive for conducting different visual tests, potentially decreasing human effort, time and costs while offering quantifiable and repeatable results instead of subjective descriptions. We validate the application of a VR system that uses only eye telemetry data to provide routine visual test measurements in a diverse population.
Methods :
We conducted a series of ophthalmic measurements using a VS Headset (Twenty/Twenty Therapeutics LLC), an eye tracking-based ophthalmic device with custom software that presents a series of time and response-dependent stimuli within the headset’s field of view while capturing eye telemetry data. These measurements allowed the generation of pupillometry, eye motility, alignment, color vision, contrast sensitivity and screening visual field parameters relying only on the user’s eye response to the presented stimuli without additional user input. The measurements for each visual test were repeated within a session or across repeated sessions for each eye to derive their repeatability and calculate their uncertainty. The success of a test for a given eye was determined by setting high standards of certainty to deem the test valid. We analyzed each test’s success rate and repeatability.
Results :
Data collected from 162 eyes from 81 subjects in routine clinical care were analyzed. Pupillometry measurements provided an intra-test variability of pupil diameter response at both dark and bright stimuli of 0.2 mm standard deviation (std). Eye motility measurements yielded an inter-test variability of range of motion for each of the six extraocular muscles within 2.4 degrees (95% confidence interval). Eye alignment measurements provided intra-test variability measuring prism diopter displacement to alternating cover/uncover response within 1.1 prism diopter std. Visual field tests yielded an intra-test variability measuring sensitivity within an outer ring neighborhood of 1.26 dB std. Color vision and contrast sensitivity measurements showed good correlation with known clinical tests.
Conclusions :
This work validates the use of a VR system to conduct a series of routine visual measurements based only on eye responses to presented stimuli. The system performs reliable measurements with high success while maintaining high reproducibility, and presents an alternative to standard visual tests while having the advantage of providing quantifiable results.
This abstract was presented at the 2023 ARVO Annual Meeting, held in New Orleans, LA, April 23-27, 2023.