C L I N I C A L

A N D

E X P E R I M E N T A L

OPTOMETRY RESEARCH PAPER

Parallel rarebits: A novel, large-scale visual field screening method Clin Exp Optom 2014; 97: 528–533 Shawn R Lin* MS BS Natalia Fijalkowski* BA Benjamin R Lin† Felix Li§ MD Kuldev Singh* MD MPH Robert T Chang* MD * Department of Ophthalmology, Byers Eye Institute at Stanford University, Stanford University School of Medicine, Palo Alto, California, United States † University of Wisconsin, Madison School of Life Sciences, Madison, Wisconsin, United States § Chinese University of Hong Kong Eye Hospital, Special Administrative Regions, Hong Kong E-mail: [email protected]

Submitted: 28 April 2013 Revised: 2 May 2014 Accepted for publication: 13 May 2014

DOI:10.1111/cxo.12221 Background: Rarebit perimetry (RBP) is a computer-based perimetric testing program with sensitivity and specificity for detection of visual field defects comparable to traditional automated perimetry. To make large-scale screening more efficient, we developed a parallel rarebit perimetric method to screen groups of subjects simultaneously. We then used this method to report the mean hit rate (MHR) among subjects aged 13 to 19 years. Methods: Rarebit perimetry was installed on computers in an existing school computer laboratory. All subjects provided medical and demographic information and underwent a basic visual examination. Testing instructions were provided to groups of up to 35 subjects and rarebit perimetry was subsequently administered. Two or three test supervisors answered questions and ensured that subjects were well aligned with their test screens. Mean hit rate, reaction times, error rates and testing time were calculated, and time estimates for rarebit, frequency doubling perimetry and Humphrey 24-2 Swedish Interactive Thresholding Algorithm (SITA) fast test were compared. Results: A total of 364 rarebit perimetric tests were conducted on 182 subjects. Of these, 154 subjects met our inclusion criteria for the reference range (three testing errors or less and visual acuity 6/9 or better). The average mean hit rate was 94.3 ± 4.63 per cent. Screening of 500 subjects using this parallel rarebit perimetric method would require approximately nine hours, which is far less than an estimated 77 hours required for frequency doubling perimetry C-20 screening tests or an estimated 127 hours required for Humphrey 24-2 SITA fast tests. Conclusion: Using our methods, rarebit perimetry can be administered in parallel to groups of subjects. The mean hit rate was comparable to that reported in previously published studies. This parallel technique may improve the efficiency of large-scale visual field screenings.

Key words: paediatric eye screening, rarebit perimetry, visual field screening

Studies in both the developed and the developing regions of the world have concluded that there is a need for a simple, costeffective visual field screening tool for the early detection of ophthalmic disease.1–5 Humphrey 24-2 fast test (H24-2) and frequency doubling perimetry (FDT) are popular techniques for screening for glaucoma but are generally too costly and too time-consuming to employ in large-scale screenings. As a result, there has been great interest in low-cost, high-sensitivity techniques to screen large groups.6 A growing body of literature supports the accuracy and reliability of rarebit perimetry (RBP) for the detection of visual field defects. Houston and colleagues7 found that rarebit perimetry was at least as sensitive as standard automated perimetry and even picked up visual field defects that were

missed by traditional perimetry. Other studies examining the use of rarebit perimetry for intracranial or ocular hypertension,8,9 post-extraction cataracts,10 amblyopia11 and homonymous hemianopsia12 have reached similar conclusions regarding the ability of rarebit perimetry to detect visual field defects. Studies have also reported that patients consider rarebit perimetry to be easier, more interesting and more comfortable than frequency doubling perimetry.13,14 In light of the emerging evidence supporting the use of rarebit perimetry for detection of visual field changes in a variety of ophthalmic conditions,15–17 we sought to develop a technique to make fast and sensitive largescale screenings possible. This pilot study describes the first published protocol for rarebit perimetry screening of groups of subjects and compares the time and cost

Clinical and Experimental Optometry 97.6 November 2014

528

required for parallel rarebit perimetry with traditional perimetry.

METHODS

Participants and preliminary examination Students from the Ngan Po Ling Secondary School in Hong Kong were offered free screening with rarebit perimetry along with follow-up for any abnormal tests at the Chinese University of Hong Kong (CUHK) eye hospital. Participation was voluntary with informed consent. This included a presentation describing the risks, benefits and alternatives to participating in the study. Groups of up to 35 subjects completed a questionnaire, providing medical histories © 2014 The Authors

Clinical and Experimental Optometry © 2014 Optometrists Association Australia

Parallel rarebits Lin, Fijalkowski, Lin, Li, Singh and Chang

for the left eye. Figure 2 provides examples of normal and abnormal rarebit perimetric screening output. Data collected in this study included mean hit rates with standard deviations, number of errors, and reaction times, as well as the total time to conduct the group screening.

Statistical analysis Figure 1. Pictorial representation of the computer laboratory screening setup for parallel rarebit perimetry. Each table held 10 computers with subjects sitting at every other computer. A string was used to mark the site where the subject’s forehead should be placed and tape on the ground marked the same location for chair positioning. A total of seven tables allowed parallel screening of 35 subjects at a time.

and demographic data. Medical histories included any personal history of diagnosis or treatment for ocular disease (excluding refractive error). Subjects were asked whether they wore glasses or contact lenses and if they had any history of ocular surgery or trauma. All subjects underwent visual acuity testing with and without pinhole using standard letter charts. Only those with visual acuity of 6/9 or better were included in the reference range.

Parallel rarebit perimetry Rarebit perimetry is a Windows software program that has emerged as a robust method for detecting subtle visual field defects. Rarebit perimetry was released in 2002 by Lars Frisén and version 4 was used in this study. Rarebit is a unique testing procedure built upon the principle that the integrity of the receptor matrix and the functionality of single receptive fields can be evaluated through the use of small stimuli (microdots). These pairs of microdots are flashed for 200 ms against a dark background, where each dot subtends half the minimum angle of resolution.18 The use of these extremely small stimuli, adjusted to the size of the receptive fields on the retina, is one of the fundamental differences between rarebit perimetry and other techniques. Subjects click a computer mouse once if they see one dot and twice if they see two dots. Based upon a subject’s reaction to multiple rounds of stimuli, the program calculates a grand mean hit rate (MHR), the overall percentage of correct responses averaged across both eyes, as well as hit rates within 24 rectangular sub-zones. The aggregate output is a map of the integrity of the

neuroretinal central visual field based upon the percentage of microdots seen by the subject.18 Care was taken to ensure that testing conditions were as close as possible to the standardised conditions described by Frisén.18 Students were tested in groups of 35 and were seated at every other computer (Figure 1). All monitors were the same model and produced 0.232-mm square picture elements (pixels) with a contrast ratio of 400:1. Stimulus luminance was calibrated to 150 cd/m2 and background luminance to 0.4 cd/m2. The ambient lighting in the room was less than 1.0 lux after darkening all windows. The full rarebit test requires subjects to perform the test at two different distances, 0.5 metres for the peripheral field and 1.0 metre for the central field. Tape distance markers were placed on the floor prior to student arrival and strings across each row were placed at the forehead level at 0.5 metres and 1.0 metre from the computer screen. Test coordinators ensured that subjects were seated at the correct distance and properly aligned in both the horizontal and vertical meridians relative to their monitor. Instructions were provided to all subjects with a PowerPoint demonstration on a projector screen. Subjects were then given approximately five minutes to practise on their own computers with the demonstration rarebit perimetric setting. One test supervisor gave directions, while test coordinators circulated the rows during the test to answer questions individually and to ensure that proper head alignment was maintained throughout the testing process. Right eyes were tested first and results were saved after each test. The test was repeated

© 2014 The Authors Clinical and Experimental Optometry © 2014 Optometrists Association Australia

Descriptive statistics, including means, medians, standard deviations, percentages and frequencies were used to summarise the data. Baseline population and subject testing characteristics were compared using frequency distributions and two-tailed t-test. Comparisons of mean hit rates, reaction times, error rates and test time among age groups were performed with nonparametric repeated analysis of variance (ANOVA). Spearman rank correlation coefficient was used to determine any association between age and mean hit rate. Linear regression analysis was used to control for error, reaction time and gender, while examining the association of age with mean hit rate. For all analyses, statistical significance was set at α less than 0.05 using two tailed t-test. Analysis was performed using Statistical Analysis Software (SAS) Enterprise 4.3 (SA institute Inc., Cary, North Carolina, USA). We certify that all applicable institutional and governmental regulations concerning the ethical use of human volunteers were followed during this research. This study had full ethics committee approval from both the Stanford School of Medicine Institutional Review Board and the Chinese University of Hong Kong (CUHK) Ethics Committee. In addition to assent obtained from students and the school, informational letters were sent home to parents. All study procedures were performed in concordance with the Helsinki Declaration.

RESULTS

Subjects One hundred and ninety-two subjects were initially offered study screening. Eight subjects did not wish to participate (4.17 per cent), leaving 184 subjects, who filled out demographic and vision history information. One subject reported a history of amblyopia and another reported a history of ocular surgery. Both of these individuals were excluded. A total of 182 subjects

Clinical and Experimental Optometry 97.6 November 2014

529

Parallel rarebits Lin, Fijalkowski, Lin, Li, Singh and Chang

95%

+20°

100% 0 0

-30°

Baseline data and results

0 0

0

0

(n = 154)

20

0

10

0

0

0

10

80

0

0

0

0

Age (years)

+30°

0

0

0

0

94%

47%

-20°

+20°

100%

0 0

% miss rates

57% 50 30

-30°

70

50 40

10

0

90

90

70

50

40

20

50

0

0

20

10

+30°

0

10 0

94%

-20°

0

0

6 (4%)

14

14 (9%)

15

42 (27%)

16

39 (25%)

17

15 (10%)

18

31 (20%)

19

7 (5%)

Gender Female (n = 69)

44.8%

Male (n = 85)

55.2%

Visual acuity (by score)

30

% miss rates

6/9 (n = 40)

Figure 2. Example of a normal (top) and abnormal (bottom) rarebit screening test result. This illustrative result was not obtained from the current study and represents an advanced superior arcuate defect with inferonasal depression. Rarebit shades boxes based on the percentage of misses within that test zone with greater shading representing a larger portion of missed stimuli. Darker areas correspond to regions in which incorrect responses are high and the circled box corresponds to the blind spot. An empty test zone indicates that the subject perceived all stimuli presented within that location. The rarebit data file also reports the mean hit rate (MHR) with standard deviation, the number of test zones with less than 90 per cent of stimuli picked up, as well as the number of errors (false-positives) during the testing period (not shown in this figure).

completed rarebit perimetric testing in both eyes (364 tests). Table 1 displays the baseline characteristics of all subjects enrolled in the study. Self-reported ethnicity in this study included: 94.1 per cent ‘Chinese,’ 4.3 per cent ‘Indian’ and 1.6 per cent reporting ‘Other.’ Among all completed rarebit perimetric tests, 95.6 per cent were conducted with three errors or less, a cut-off deemed necessary to test reliability based upon previous studies.18 Poor visual acuity (worse than 6/9 in one or both eyes) was the primary reason for exclusion with 20 subjects failing to meet the visual acuity criterion. Among these, 14 reported not bringing their glasses or wearing their contact lenses. One hundred and fifty-four subjects (80.2 per cent) comprising 308 eyes met the inclusion criteria and were included in the final

analysis for our reference range. The mean age of all subjects in this group was 16.0 years, ranging from 13 to 19 years (average (mode) 16 years). Females comprised 44.8 per cent of the study group. The average mean hit rate was 93.9 ± 5.28 per cent in the right eye and 94.7 ± 3.88 per cent in the left eye with a grand mean hit rate of 94.3 ± 4.63 per cent for all reference subjects. Table 2 shows a frequency distribution for the rectangular sub-zones. There was no statistically significant difference between the mean hit rates of the two eyes (p = 0.48), indicating an absence of meaningful short-term learning effects. In addition, there was no significant difference between the mean hit rate for the groups with visual acuity of 6/6 and 6/9 (p = 0.24). Figure 3 compares the results for the mean hit rate obtained in this study to that of

Clinical and Experimental Optometry 97.6 November 2014

530

13

6/6 (n = 114) 90%

Number (%) subjects

74.0% 26.0%

Grand mean hit rate

94.3 ± 4.63%

Right eye

93.9 ± 5.28%

Left eye

94.7 ± 3.88%

Errors

0.67 ± 0.89

Test time (seconds)

256.1 ± 31.5

Reaction time (seconds)

0.43 ± 0.06

Table 1. Baseline data and parallel rarebit perimetric results for all subjects who met inclusion criteria. Inclusion criteria for reference range database included visual acuity better than or equal to 6/9 bilaterally and three or fewer errors on the test.

previously published rarebit perimetric studies. Although the studies included various age ranges, our mean hit rate and standard deviations were similar to those reported in other studies. Analysis by gender revealed no statistical difference in mean hit rate (p = 0.12), errors (p = 0.18) or test time (p = 0.24). Interestingly, there was a statistically significant, yet clinically minimal difference in reaction times between female (mean 0.48 seconds) and male subjects (mean 0.36 seconds) (p = 0.02). The overall mean reaction time was 0.43 ± 0.06 seconds during each test. Spearman rank coefficient confirmed that there was no association between age and mean hit rate in the study population (r = -0.02, p = 0.46). Linear regression analysis demonstrated that after controlling for gender, number of errors and visual acuity, © 2014 The Authors

Clinical and Experimental Optometry © 2014 Optometrists Association Australia

Parallel rarebits Lin, Fijalkowski, Lin, Li, Singh and Chang

Sub-zone

examined the eight right eyes with the lowest mean hit rates (less than 85 per cent) by reviewing mean hit rate performance in the fellow eye. Interestingly, the right eye mean hit rate was worse than the left eye in all of these cases. As the right eye was the first eye we tested, this suggests that a learning curve may have contributed to a low right eye mean hit rate for these eight patients.

Overall zone

mean hit rate (%)

frequency (%)

100%

81%

90%

10%

80%

6%

70%

2%

60%

1%

50% or less

0%

Screening time and cost Time estimates for a theoretical group of 500 subjects were calculated for parallel rarebit perimetry, frequency doubling perimetry C-20 screening test and Humphrey 24-2 Swedish Interactive Thresholding Algorithm fast test screenings using testing times from Wadood and colleagues.19 Subject testing time was calculated for three testing modalities by estimating time necessary for test set-up, equipment transportation and subject consent and training. Table 3 describes the total time and cost for each modality. When compared to previously reported perimetric techniques, time required for screening with parallel rarebit perimetry during this study was significantly less: with an estimated nine hours required to screen 500 subjects with parallel rarebit perimetry, compared to 77 hours using sequential frequency doubling perimetry and 127 hours using sequential Humphrey. Rarebit perimetric software was free and installed on computers already present at the school computer laboratory. The cost of infrastructure should be acknowledged, although this is not directly included in costaccounting analysis. This is compared to the cost of traditional perimetry, which requires equipment (-$6,000 for frequency doubling perimetry, -$25,000 for Humphrey), trained technicians and transportation.

Table 2. Rarebit perimetry sub-zones frequency. There are 24 total sub-zones, with one sub-zone overlapping the blind spot and the remaining 23 zones averaged to calculate the mean hit rate. As shown, 81 per cent of zones had a hit rate of 100 per cent over the entire study. Chin21 n=54

86.3% ± 14

Salvetat16 n=53

87.6% ± 6.5

Brusini27 n=41

88.6% ± 4.8

Salvetat17 n=75

91% ± 5.7

This study n=182

94.3% ± 4.6

Martin14 n=54

95.5% ± 2.7

Hellgren23 n=52

96%

Corallo9 n=30

96.2% ± 2.0

Aleci12 n=14

96.8% ± 1.75 0

10 20 30 40 50

60 70 80 90 100

DISCUSSION

Mean hit rate % Figure 3. Literature comparison of mean hit rates and standard deviations for normal subjects. The results obtained using our parallel rarebit perimetry are similar to those obtained in other studies with individual subject testing. there was no statistically significant effect of subject age on mean hit rate (p = 0.69). It should be acknowledged that a mean hit rate instrumental ceiling affects statistical analysis. The variance in our sample is artificially reduced because many students scored close to the instrument ceiling of 100 per cent

mean hit rate, In the case of our study, this means that the differences between comparison groups for gender, age et cetera are underestimated as a result of this constrained variance. Figure 4 shows the mean hit rate in the right eye plotted against age. We further

© 2014 The Authors Clinical and Experimental Optometry © 2014 Optometrists Association Australia

Parallel rarebit perimetry is a practical and effective large-scale screening method in a computer-proficient population. Although organisations likely to employ rarebit perimetry would also require a technician to set up the test and provide training, this upfront investment is offset by the higher throughput of the parallel examination. In addition, rarebit perimetry is free of charge and can be installed on any computer worldwide. Standard screenings with frequency doubling perimetry require specialised, proprietary equipment that is expensive and often difficult to obtain.

Clinical and Experimental Optometry 97.6 November 2014

531

Parallel rarebits Lin, Fijalkowski, Lin, Li, Singh and Chang

100% 95%

Mean hit rate (%)

90% 85% 80% 75% 70% 65% 60% 13

14

15

16

17

18

19

20

Age (years)

Figure 4. Plot of right eye mean hit rate versus age. Spearman rank coefficient confirmed that there was no association between age and mean hit rate in the study population (r = -0.02, p = 0.46). Linear regression analysis demonstrated no statistically significant effect of subject age on mean hit rate (p = 0.69). Martin and Nilsson20 demonstrated in a small cohort of children that rarebit perimetry is a sensitive and specific way to detect peripheral visual field defects in paediatric populations. Other studies have demonstrated that rarebit perimetry is repeatable and reliable,21 with less retest variability than traditional perimetry.22 Some studies14,21,23 have suggested that because rarebit perimetry is similar to a computer game, children actually have fewer fixation losses and fewer false-positives with rarebit when compared with traditional perimetry. Furthermore, patients in these studies reported greater satisfaction with rarebit perimetry.14 Using parallel rarebit perimetry, we were able to measure a mean hit rate of 94.3 ± 4.63 per cent for the 13 to 19 age group with significantly less time and cost expenditure than would have been required with traditional perimetry. This mean hit rate was comparable to other paediatric study populations.14,20 The results of our pilot study support the feasibility of the parallel rarebit perimetry. We acknowledge that screening for field defects in older populations would yield more abnormals compared to paediatric screenings and that technological familiarity could be a challenge with older adults. In these cases, it may be prudent to reduce the screening group size from 35 to a more manageable number. This would help maintain cost and efficiency benefits and would allow for increased supervision and assistance.

We also believe that refinement in test administration would improve results. It can be difficult to ensure standardised conditions for the test, especially in regard to monitor pixel size and contrast ratios. Stability of these parameters is crucial to the accuracy of the rarebit perimetry, as discussed by Nilsson, Wanger and Martin.24 In addition, it is important to ensure that subjects maintain the correct screen alignment and are not distracted during the test. This requires attentive proctoring, albeit not necessarily by trained technicians. Simple measures could be taken to further optimise the testing environment. For example, headphones for each testing station could minimise distraction, as students were prone to giggling when a fellow participant elicited an error noise from his examination. Future studies could also use chin-rests to ensure subjects are accurately positioned with less supervision. Nevertheless, the subjects were cooperative and generally did not have trouble with the test. Finally, our study was conducted in Hong Kong and the subjects were predominately (more than 90 per cent) Chinese. This population was initially selected because of the potential for higher rates of significant myopic field defects among subjects of Asian descent, where peripheral visual field screening modalities could be particularly beneficial.25,26 The test was conducted in English and teachers were present to translate any questions that the proctors could

Clinical and Experimental Optometry 97.6 November 2014

532

not answer. Although there may be some concern that students did not understand all directions, the generally low error rate in this study demonstrates that students appropriately completed the test. Rarebit’s mean hit rate does not have a well-established normative database and the outcome or mean hit rate is not yet as recognised by clinicians as frequency doubling perimetry and Humphrey results. However, by aggregating multiple studies, one can reasonably infer that a mean hit rate cutoff under 80 per cent may indicate a need for further ophthalmic examination. Prior research has demonstrated that control subjects for various diseases, including glaucoma,16,27,28 amblyopia,11 optic neuropathy,20,29 empty sella syndrome30 and idiopathic intracranial hypertension8 have mean hit rates above 80 per cent, while subjects with these conditions generally demonstrate a mean hit rate below 80 per cent, with a statistically significant difference in all the above-mentioned studies. Furthermore, normative database work in adults focused on optimising receiver-operator curves (ROC) for screening with rarebit perimtery has reported 80 per cent mean hit rate as the best cut-off for optimising the sensitivity and specificity of rarebit perimetry for detection of visual field defects;18,27 however, this does not take into account the fact that a focal defect could be masked by an overall mean hit rate above 80 per cent. In these cases, it would be prudent to examine each subject’s screening test result (Figure 2) to ensure that no focal defects are missed. This cutoff should also be calibrated for age, as studies have shown a predictable decrease in mean hit rate in older patients.21 To define a better reference range for this population, it would be prudent to select a more even age distribution and perform additional testing. Rarebit perimetry could fill the need for a free screening program to detect visual field defects requiring further ophthalmic examination. In addition to the school computer laboratory screening model described in this study, the technology may even be adaptable for home self-testing or to help monitor patients diagnosed with disease. We plan to replicate this study in older adults with follow-up examinations as the next step to testing this procedure. Although future studies are undoubtedly required to validate this approach, parallel rarebit perimetry could make it more feasible to conduct large-scale visual field screening to detect early glaucomatous change. © 2014 The Authors

Clinical and Experimental Optometry © 2014 Optometrists Association Australia

Parallel rarebits Lin, Fijalkowski, Lin, Li, Singh and Chang

Parallel rarebit Cost

Free

Frequency doubling (C-20)

Humphrey (24-2 SITA)

Equipment (∼$6,000 / unit)

Equipment (∼$25,000 / unit)

Trained technician

Trained technician

Transportation

Transportation

1 hour

1 hour

Equipment transportation

Onsite computer room

Equipment setup / calibration

2 hours

1 hour

1 hour

Subject consent / registration

10 minutes for 35 subjects

5 minutes for each subject

5 minutes for each subject

Subject training

10 minutes for 35 subjects

2 minutes for each subject

2 minutes for each subject

Subject testing

10 minutes for 35 subjects

2 minutes for each subject

8 minutes for each subject

Total time to screen 500 subjects

9 hours

77 hours

127 hours

Table 3. Comparison of the theoretical costs and calculated time required for frequency doubling perimetry, Humphrey and rarebit perimetric modalities to screen 500 subjects. Time required for traditional standard automated perimetry determined from previous published data in Wadood and colleagues.19 ACKNOWLEDGEMENTS

We would like to thank the Ngan Po Ling Secondary School in Hong Kong, SAR and Saint Theresa Secondary School in Hong Kong, SAR for their support. We are grateful to the Chinese University of Hong Kong (CUHK) Eye Hospital with special thanks to Dr Fan and Dr Tam for all their support. We also thank Lam Siu Chuen, without whom this work would not have been possible. This research was supported by a Stanford Travelling Scholars Research Grant, the Stanford School of Medicine, Medical Scholars Research Program and an American Glaucoma Society MAPS Grant. REFERENCES 1. Congdon NG, Friedman DS, Lietman T. Important causes of visual impairment in the world today. JAMA 2003; 290: 2057–2060. 2. Munoz B, West SK. Blindness and visual impairment in the Americas and the Caribbean. Br J Ophthalmol 2002; 86: 498–504. 3. Rahi JS. Childhood blindness: a UK epidemiological perspective. Eye (Lond) 2007; 21: 1249–1253. 4. Salomao SR, Mitsuhiro MR, Belfort R Jr. Visual impairment and blindness: an overview of prevalence and causes in Brazil. An Acad Bras Cienc 2009; 81: 539–549. 5. Uzma N, Kumar BS, Khaja Mohinuddin Salar BM, Zafar MA, Reddy VD. A comparative clinical survey of the prevalence of refractive errors and eye diseases in urban and rural school children. Can J Ophthalmol 2009; 44: 328–333. 6. Schiefer U, Gisolf AC, Kirsch J, Selbmann HK, Zrenner E. [Noise field screening. Results of a television field study for detection of visual field defects]. Ophthalmologe 1996; 93: 604–616. 7. Houston SK, Weber ED, Koga SF, Newman SA. Rarebit perimetry for bedside testing: comparison with standard automated perimetry. J Neuroophthalmol 2010; 30: 243–247.

8. Celebisoy N, Ozturk T, Kose T. Rarebit perimetry in the evaluation of visual field defects in idiopathic intracranial hypertension. Eur J Ophthalmol 2010; 20: 756–762. 9. Corallo G, Iester M, Scotto R, Calabria G, Traverso CE. Rarebit perimetry and frequency doubling technology in patients with ocular hypertension. Eur J Ophthalmol 2008; 18: 205–211. 10. Nilsson M, Abdiu O, Laurell CG, Martin L. Rarebit perimetry and fovea test before and after cataract surgery. Acta Ophthalmol 2010; 88: 479–482. 11. Agervi P, Nilsson M, Martin L. Foveal function in children treated for amblyopia. Acta Ophthalmol 2010; 88: 222–226. 12. Aleci C, Usai T. Testing spatial detection and light sensitivity in homonymous hemianopia by rarebit and conventional automated perimetry. Open Ophthalmol J 2008; 2: 153–159. 13. Martin L, Wanger P. New perimetric techniques: a comparison between rarebit and frequency doubling technology perimetry in normal subjects and glaucoma patients. J Glaucoma 2004; 13: 268– 272. 14. Martin L. Rarebit and frequency-doubling technology perimetry in children and young adults. Acta Ophthalmol Scand 2005; 83: 670–677. 15. Alencar LM, Medeiros FA. The role of standard automated perimetry and newer functional methods for glaucoma diagnosis and follow-up. Indian J Ophthalmol 2011; 59 Suppl: S53–S58. 16. Salvetat ML, Zeppieri M, Tosoni C, Parisi L, Brusini P. Non-conventional perimetric methods in the detection of early glaucomatous functional damage. Eye (Lond) 2010; 24: 835–842. 17. Salvetat ML, Zeppieri M, Parisi L, Brusini P. Rarebit perimetry in normal subjects: test-retest variability, learning effect, normative range, influence of optical defocus, and cataract extraction. Invest Ophthalmol Vis Sci 2007; 48: 5320–5331. 18. Frisen L. New, sensitive window on abnormal spatial vision: rarebit probing. Vision Res 2002; 42: 1931–1939. 19. Wadood AC, Azuara-Blanco A, Aspinall P, Taguri A, King AJ. Sensitivity and specificity of frequencydoubling technology, tendency-oriented perimetry and Humphrey Swedish interactive threshold algorithm-fast perimetry in a glaucoma practice. Am J Ophthalmol 2002; 133: 327–332.

© 2014 The Authors Clinical and Experimental Optometry © 2014 Optometrists Association Australia

20. Martin LM, Nilsson AL. Rarebit perimetry and optic disk topography in pediatric glaucoma. J Pediatr Ophthalmol Strabismus 2007; 44: 223–231. 21. Chin CF, Yip LW, Sim DC, Yeo AC. Rarebit perimetry: normative values and test-retest variability. Clin Experiment Ophthalmol 2011; 39: 752–759. 22. Vislisel JM, Doyle CK, Johnson CA, Wall M. Variability of rarebit and standard perimetry sizes I and III in normals. Optom Vis Sci 2011; 88: 635–639. 23. Hellgren K, Hellstrom A, Martin L. Visual fields and optic disc morphology in very low birthweight adolescents examined with magnetic resonance imaging of the brain. Acta Ophthalmol 2009; 87: 843–848. 24. Nilsson M, Wanger P, Martin L. Perception of very small visual stimuli in the fovea: normative data for the Rarebit Foveal Test. Clin Exp Optom 2006; 89: 81–85. 25. Ohno-Matsui K, Shimada N, Yasuzumi K, Hayashi K, Yoshida T, Kojima A, Moriyama M et al. Longterm development of significant visual field defects in highly myopic eyes. Am J Ophthalmol 2011; 152: 256–265. 26. Kumar RS, Baskaran M, Singh K, Aung T. Clinical characterization of young Chinese myopes with optic nerve and visual field changes resembling glaucoma. J Glaucoma 2012; 21: 281–286. 27. Brusini P, Salvetat ML, Parisi L, Zeppieri M. Probing glaucoma visual damage by rarebit perimetry. Br J Ophthalmol 2005; 89: 180–184. 28. Hackett D, Anderson, A. Determining mechanisms of visual loss in glaucoma using rarebit perimetry. Optom Vis Sci 2011; 88: 48–55. 29. Frisen L. Performance of a rapid rarebit centralvision test with optic neuropathies. Optom Vis Sci 2012; 89: 1192–1195. 30. Yavas GF, Kusbeci T, Eser O, Ermis SS, Cosar M, Ozturk F. A new visual field test in empty sella syndrome: rarebit perimetry. Eur J Ophthalmol 2008; 18: 628–632.

Clinical and Experimental Optometry 97.6 November 2014

533

Copyright of Clinical & Experimental Optometry is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.

Parallel rarebits: a novel, large-scale visual field screening method.

Rarebit perimetry (RBP) is a computer-based perimetric testing program with sensitivity and specificity for detection of visual field defects comparab...
267KB Sizes 1 Downloads 5 Views