Development of Automated Visual Acuity Measurement Using a Calibration-Free Eye-Tracking System.

Purpose Infant vision assessment often relies on grating acuity; however, its objectivity and convenience must be improved. A calibration-free eye-tracking system, even in preverbal children, enables easy and precise gaze analysis. This pilot study aimed to develop a reliable automated monocular vision screening. Methods Participants (n=118) underwent a grating visual acuity test using the eye-tracking system. Correlations between the grating acuity, uncorrected visual acuity, and refractive error were analyzed across different cutoff values of fixation duration percentage. Results Strong correlations were found between the grating acuity and refractive error at 69% and 88% thresholds. Similar correlations with uncorrected visual acuity were noted at 70% and 89% thresholds. False-negatives around the 70% threshold were noted, indicating potential overestimation of acuity in cases of low visual acuity/high refractive error. Discussion The results highlight the feasibility of calibration-free eye-tracking system-based monocular vision screening with an optimal screening threshold of 90%.

Measurement Accuracy When Using Spot Vision Screener With or Without Cycloplegia in Young Adults.

Purpose: There are many unclear points about the accuracy of measurement of cycloplegic refraction using the Spot Vision Screener (SVS). This study aimed to investigate the accuracy of SVS measurements with cycloplegia for myopia. Materials and Methods: Forty-nine healthy subjects were included, and refraction was measured. Objective refractions were measured by SVS, table-mounted autorefractometer (RT7000), and handheld autorefractometer (Retinomax Screeen) at noncycloplegic and cycloplegic conditions by 1% cyclopentolate. Subjective noncycloplegic refraction was obtained by a visual acuity and refraction test performed by certified orthoptists using a cross-cylinder. One-way repeated-measures analysis of variance was used to examine whether the measured refractions fluctuate due to different reflection tests. Results: In the noncycloplegic condition, the mean (±standard deviation) spherical equivalent (SE) measured by subjective method, SVS, RT7000, and Retinomax Screeen were -2.56 ± 3.00, -2.62 ± 2.38, -3.05 ± 2.84, and -3.26 ± 2.97, respectively. The subjective SE and objective SE measured by SVS had significantly less myopic value than the objective SE measured by two autorefractometers (p < 0.001). In the cycloplegic condition, the mean (± standard deviation) SE measured by SVS, RT7000, and Retinomax Screeen were -2.07 ± 2.66, -2.62 ± 2.98, and -2.66 ± 3.02, respectively. The objective SE measured by SVS had significantly less myopic value than SEs measured using other methods (p < 0.001). In the cycloplegic condition, SVS showed a fixed error wherein the SE was more hyperopic than that with the subjective method and SVS had a proportional error. Conclusion: In the measurement under cycloplegic conditions, use of an autorefractometer rather than a photorefractometer such as SVS was preferable.