Abstract
Purpose :
Currently, there are no standardized methods to select, quantify or train eye movements in patients with central vision loss (CVL). We developed 2 eccentric viewing (EV) training systems, in-person and remote, to implement and compare alternative feedback methods for CVL rehabilitation.
Methods :
CVL patients (n=8 in-lab (median age=71years ± 20 standard deviation); n=8 remote (61 ± 11); n=9 normally-sighted controls (28 ± 5)) were recruited. The remote system, consisting of a laptop and eye tracker (Tobii EyeX or GazePoint GP3), was mailed to participants. Matlab speech and an experimenter on Zoom guided participants through the remote experiment. The in-lab system was led by a researcher and utilized an Eyelink eye tracker. In both systems, a PRL in the inferior visual field was trained with audio, contracting ring, gaze-contingent ring, image of a hand, and tactile feedback in random order and were compared to free viewing. The contrast sensitivity function (CSF) at the PRL was assessed via bandpass-filtered HOTV letters, updated using an adaptive algorithm; fitted after 25 correct responses; maximum 50 trials per condition.
Bivariate-Contour-Ellipse-Areas (BCEAs), Area-Under-the-CSF (AUCs), and CSF Acuities were calculated to measure oculo-motor and perceptual performance, respectively. Analysis-of-variance models and planned multiple comparisons were used to determine statistical effects.
Results :
We were able to measure the CSF and BCEAs of all subjects both in-lab and remotely, however, eye tracking data loss was higher in remote-CVL patients, likely due to EV. BCEA was significantly reduced in all feedback conditions in the remote-control group (p=0.003), but not for the in-lab control group. In all groups, there were no significant differences in Acuity or AUC across feedback mechanisms.
Conclusions :
We developed a novel remote EV training system, which successfully measured contrast-sensitivity in CVL patients and successfully tracked eyes in a subset of participants. Although the present feedback methods did not affect the visual function endpoints tested in this study, this system may be used to evaluate other rehabilitation approaches and functional outcomes.
This abstract was presented at the 2022 ARVO Annual Meeting, held in Denver, CO, May 1-4, 2022, and virtually.