Skip to Main Content
 

Global Search Box

 
 
 

ETD Abstract Container

Abstract Header

Computer-assisted Adaptive Methods of Measuring Visual Acuity

Andrews, Erin Jessica
http://rave.ohiolink.edu/etdc/view?acc_num=osu1492549443966615

Abstract Details

2017, Master of Science, Ohio State University, Vision Science.
Visual acuity measurement has evolved very little throughout the course of its history. Few changes have been proposed and even fewer have been adopted to advance the process of measuring visual ability. One major and widely accepted advancement includes logMAR size progression (Bailey-Lovie and Early Treatment Diabetic Retinopathy Study) chart designs. Logarithmic progression allows for visual acuity measurements to be expanded over a larger range of values, and it also provides the opportunity to quantify acuity changes in terms of “lines of change”. Logarithmic progression also allows for easy letter-by-letter scoring, which is a more accurate and repeatable determination of visual acuity when compared to line-by-line scoring. Additional investigations to improve visual acuity measurement testing have aimed to increase reliability, repeatability, and efficiency. For example, computerized methods of visual acuity testing have been tested. These methods have faced several challenges. Further, adaptive psychophysical methods such as BestPEST and ZEST have been employed in some investigations to improve efficiency and accuracy of threshold visual acuity testing. Adaptive procedures manipulate testing based on previous answers and concentrate testing close to the estimated threshold to efficiently determine an accurate and reliable end point measurement of threshold. Both computer-assisted and adaptive procedures for measuring visual acuity still require investigation and improvement. Visual acuity measurement is one of the most commonly used and most important measures of visual quality and function, and is likely the most commonly used measurement in clinical optometry. Therefore, it is important for the measurement to be standardized and efficient, and its scoring to be accurate, reliable, and repeatable. This study aims to investigate test and scoring procedures in adaptive computer-assisted visual acuity. Computer-assisted VA procedures were evaluated in 52 normally-sighted subjects divided into three experiments. LogMAR procedures were compared to three versions of adaptive methods (PEST, ZEST, and curve fits to the probability of seeing). Repeatability was evaluated as test-retest (TRT) distributions, separated by 1-2 weeks, and by “simultaneous” scores, i.e. trials from test and retest randomly interleaved within the same session. Subjects responded verbally or by typing. LogMAR acuity was measured with a standard logMAR chart, and with a modified single-letter method. Finally, a response-verification procedure required subjects to confirm letter selection in order to minimize typing errors. Test times, mean visual acuity, and test-retest differences were evaluated and compared for all experiments. Within each experiment, the three adaptive scoring methods produced very similar results, with all differences between curve-fitting, PEST and ZEST being less than 0.02 log units, or less than one “letter” of acuity. The method used to run each experiment (CF for Expt 1, ZEST for Expts 2 and 3) also produced the smallest test-retest differences in the final score. Confidence intervals for the between-session conditions were slightly wider than values found in the literature for similar protocols. Within-session TRT values were smaller than between-session values for all experiments, with more than half the variance in between-session repeatability attributed to actual differences in acuity between sessions. The response verification procedure did not produce narrower within-session test-retest distributions compared to other procedures without the required letter confirmation. Within-session repeatability is substantially better than repeatability from sessions separated by days or weeks. Repeatability from sessions separated by days or weeks may be attributed to true visual acuity change over time, whereas repeatability within a session may represent the variability of the testing procedure itself. Adaptive, computer-assisted methods have promise in improving the efficiency and repeatability of VA testing.
Thomas Raasch, O.D, PhD (Advisor)
83 p.
Health Sciences; Optics
Visual acuity measurement; BestPEST; ZEST; clinical optometry; adaptive computer-assisted visual acuity

Recommended Citations

Citations

  • Andrews, E. J. (2017). Computer-assisted Adaptive Methods of Measuring Visual Acuity [Master's thesis, Ohio State University]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=osu1492549443966615

    APA Style (7th edition)

  • Andrews, Erin. Computer-assisted Adaptive Methods of Measuring Visual Acuity. 2017. Ohio State University, Master's thesis. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=osu1492549443966615.

    MLA Style (8th edition)

  • Andrews, Erin. "Computer-assisted Adaptive Methods of Measuring Visual Acuity." Master's thesis, Ohio State University, 2017. http://rave.ohiolink.edu/etdc/view?acc_num=osu1492549443966615

    Chicago Manual of Style (17th edition)

osu1492549443966615
623
© 2017, all rights reserved.
This open access ETD is published by The Ohio State University and OhioLINK.