Abstract
Purpose :
Technological developments have resulted in the availability of numerous head-worn assistive devices (HMDs) for people with vision loss (PVL). The specifications for these devices, provided by companies, include large nominal magnification and contrast enhancements. However, the actual values and the resulting performance gains from these settings have yet to be quantified in a clinical setting. In the present study, we measured the gains in performance using three HMDs: eSight4, Eyedaptic EYE3, and IrisVision Inspire.
Methods :
We did not measure contrast directly from the HMD displays; instead, we used psychophysical methods as a surrogate for assessing contrast gain. Four control subjects with typical contrast sensitivity (CS) and 15 PVL participated in this study. A MARS letter contrast chart was placed at a viewing distance of 40 cm, and the subjects read in decreasing contrast order until they could no longer recognize the letters. The log CS value of the last letter read correctly was recorded. This was done at baseline with habitual near correction and while wearing each HMD with nominal contrast levels set at 0,10, 25, 50, 75, or 100.
Results :
Log CS without HMDs averaged 1.68 (~2% contrast) for the control subjects; however, when using these devices, log CS was less than baseline at all nominal contrast settings (F = 31.12, P < 0.001). For the PVL, there was a median increase in contrast threshold of 19% (median nominal contrast setting = 15) when using eSight, 5% (median nominal contrast setting = 50) for Eyedaptic EYE 3, and a decrease in CS of 3% % (median nominal contrast setting = 54) when using IrisVision Inspire (F= 6.56, P = 0.005). There was a negative relationship between the change in contrast thresholds with the devices and baseline contrast thresholds (F = 9.9, P = 0.003).
Conclusions :
Image contrast manipulations in electronic devices differ from traditional psychophysical-directed changes. When using the devices, the contrast of low-contrast targets (< 0.5) decreases and quickly saturates at zero percent, and that of higher contrast targets (>0.5) rapidly approaches one as higher nominal contrast gain values are selected. Therefore, the gain in CS that can be expected from digital image contrast manipulation depends upon baseline CS.
This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.