Purchase this article with an account.
Anshul Gupta, Stephen Engel, Erik J Van Kuijk, Aurelie Calabrese, Jacob Sanders, Arthur Erdman, Gordon E Legge; Evaluating Reading Performance with a Head Mounted Aid for Central Visual Field Loss. Invest. Ophthalmol. Vis. Sci. 2016;57(12):4430.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
We developed and tested a head-mounted, eye-tracker based image enhancement aid for central field loss (CFL). The device remaps the visual information lost to a simulated central scotoma in real time, and displays it on a head mounted display’s (HMD) screen. We used it to test the hypothesis that remapping lost information onto a functional retinal location improves reading speed.
The device was comprised of a Sensics zSight HMD, with a 60° diagonal field of view, fitted with a monocular Arrington Research 60Hz eye tracker. The eye tracker provided gaze location and enabled implementation of simulated central scotomas. The images of text were remapped using a Gaussian Bump algorithm. Remapped images were displayed on the HMD screen in real time. Artificial scotomas subtending 4° and 8° were simulated on 7 normally sighted subjects, and reading speeds with and without remapping were measured. For each condition, subjects silently read 7 MNREAD sentences with x-height of 0.83°. The line spacing was set to 1.245° (1.5x). Reading speed was normalized by baseline speed without a scotoma. Eye tracker calibration was assessed in real time, and if calibration was lost, the trial was excluded.
For the 8° scotoma with no remapping, subjects completed 13 of the 42, or 31% of the valid sentence trials, with an average reading speed of 23% (± 1.6%) of their speed without a scotoma. With remapping, subjects completed 46 of the 49, or 94% of the sentences presented, with an average reading speed of 39% (±11.6%) of their speed without a scotoma. The benefits of remapping are clearly indicated by the large increase in completion rate and reading speed. For the 4° scotoma, all subjects completed all trials. The average normalized reading speed was 67% (±7%) of baseline with remapping and 67% (±13%) without remapping. Remapping had no significant effect on reading speed for this scotoma size.
Testing with normally sighted subjects with simulated CFL suggests that the remapping device shows promise in improving reading speeds in CFL patients, especially those with larger scotomas. More testing is underway, and a future goal is to test on CFL patients.
This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.
Artificial Scotoma with and without Remapping
This PDF is available to Subscribers Only