Resolution perimetry using Landolt C |
239 |
|
|
RESOLUTION PERIMETRY USING LANDOLT C
HIROSHI YAKUSHIGAWA, YASUHIRO NISHIDA, TAICHIRO MIYAKE and
KAZUTAKA KANI
Department of Ophthalmology, Shiga University of Medical Science, Shiga, Japan
Abstract
We developed a computer program to measure acuity thresholds at different eccentricities using Landolt C targets. The subjects were told to gaze at a fixation target in the center of a notebook computer monitor. Subjects responded by pressing a key signifying the direction of the gap in the target or, when it could not be resolved, by a ‘don’t know’ key. Targets were presented in random positions for 200 msec. The examination was performed using the constant stimuli method, where the Landolt C, in each of seven different sizes, was randomly displayed 30 times at each eccentricity. The probability of a correct answer in each different size was plotted at each eccentricity. Probability-of-seeing curves were drawn, and then acuity threshold determined. This system makes it easy to measure acuity thresholds at different eccentricities, and can be used to evaluate the parvocellular visual system.
Introduction
Measurement of acuity thresholds at different eccentricities is one way to evaluate peripheral visual function. Over a century ago, it was established1 that acuity decreased with eccentricity. Green2 measured eccentric acuity thresholds using interference fringes formed directly on the retina by a neon-helium gas laser. Several other studies have used Landolt C targets3-6 for the assessment of eccentric acuity thresholds. However, it seemed that not only the equipment, but also the testing protocols were too complicated to be used as a clinically measure. In recent studies,7,8 test targets were displayed automatically using a personal computer. We developed a new type of perimeter, which is able to easily measure eccentric acuities using Landolt C targets displayed on a notebook computer monitor.
Address for correspondence: Hiroshi Yakushigawa, MD, Department of Ophthalmology, Shiga University of Medical Science, Seta, Otsu, Shiga 520-2192, Japan. Email: yakushi@belle.shiga-med.ac.jp
Perimetry Update 2002/2003, pp. 239–243
Proceedings of the XVth International Perimetric Society Meeting, Stratford-upon-Avon, England, June 26–29, 2002
edited by David B. Henson and Michael Wall
© 2004 Kugler Publications, The Hague, The Netherlands
240 |
H. Yakushigawa et al. |
Methods
Equipment
The program was developed using Visual Basic® 6.0 in the Windows® 2000 (Microsoft Corporation, USA) operating environment. The targets were generated on a notebook computer with 14.1-inch TFT XGA 1024×768 color LCD monitor with a separate keypad for the patient to respond (Figs. 1 and 2). The background color on the monitor
Fig. 1. The notebook computer with the separate keypad used in this examination. A black X-shaped fixation target and Landolt C are displayed on the monitor.
Fig. 2. The keypad is used by the subject to indicate the direction of the open-space in the Landolt C target (up, down, right or left). When the subjects cannot see the open-space, they push the center button with a question mark.
Resolution perimetry using Landolt C |
241 |
was white and the brightness was 250 apostilb. There was always a black X-shaped fixation target in the center of the monitor. The test target was a black Landolt C with the open-space in the up, down, right or left position. The examination was carried out in a bright room, and the contrast between the background and the target was 1:40. The distance between the subject and target was 5.0 meters. The measurement was conducted according to the constant stimuli method as mentioned in the following testing protocol.
Testing protocol
The subjects examined were three males with no ocular diseases, and with corrected vision of over 1.5. The examination was performed in the fully-corrected dominant eye. The target eccentricities were zero, 0.5, 1.0, 2.0, 5.0 and 10 degrees along the nasal superior retina for subject A, and the horizontal meridian of the temporal retina for subjects B and C. Landolt Cs of 17 different sizes were prepared: i.e., 0.1, 0.12, 0.15, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0, 1.2, 1.5, 1.75, 2.0, and 2.25. Seven sequential Landolt Cs in different size were selected from 0.1 to 2.25 on the basis of eccentricity. The computer was programmed to randomly display the Landolt C 30 times in each of seven different sizes at each eccentricity. The target exposure time was 200 msec, and the next target was presented after the subject responded. The subjects were told to gaze at the fixation target, and answer with the direction of the displayed Landolt C by means of pushing the keypad button for either up, down, right, left or question mark.
Results
The probability plots of correct answers for each target size at six different eccentricities for subject A are shown in probability-of-seeing curves (Fig. 3a). Probability plots at each eccentricity show gaussian distribution (normal distribution), and good linear fit when plotted on a normal probability paper (Fig. 3b). The eccentric acuity threshold was defined by the 50% probability value on the graph in Figure 3b. The eccentric
Subject A, 63-year-old male
Fig. 3a. Probability-of-seeing curves at six different eccentricities, from 0-10 degrees, for subject A.
242 |
H. Yakushigawa et al. |
Fig. 3b. Probabilities at each eccentricity show gaussian distribution on normal distribution paper. Each eccentric acuity is determined by the target size at the 50% probability.
63-year-old male
43-year-old male
39-year-old male
Fig. 4. Eccentric acuity for the three subjects are shown together with the result reported by Green.2
Resolution perimetry using Landolt C |
243 |
acuities for three subjects are shown in Figure 4. Acuity at zero degrees in eccentricity is from 1.5-1.75, which is almost the same as that found for distance with this type of target. Acuity at 0.5 degrees in eccentricity is from 1.3-1.5, which is almost 85% of the vision at zero degrees. On the other hand, acuity at 10 degrees in eccentricity is from 0.2-0.38, and the differences among the three subjects are the largest. The mean examination time for this strategy was about 30 minutes. No significant difference was shown in eccentric acuity if the exposure time was increased from 200-500 msec. No significant differences in eccentric acuity were found between the nasal and temporal retina, at 0.5, 1.0, 3.0, 5.0 and 10 degrees in eccentricity, for subject C.
Discussion
Our results show that the probability plots for each target size, at each eccentricity, show a gaussian distribution, and the eccentric acuity agrees with the published values reported by Green.2 The acuity in the present study can be more precisely determined than is normal with visual acuity measures because the constant stimuli method is employed. However, the constant stimuli method takes a long time to perform and requires perseverance and concentration from the subject and examiner. In the present study, the computer made it much easier to measure eccentric acuity. The exposure time of the Landolt C was set at 200 msec, in order to prevent saccadic eye movement influencing the results (reaction times being >200 msec).9-11 The mean examination time for acuity measurements was similar to that of a threshold test using a computerassisted perimeter. The measurement distance was 5.0 meters. If the distance was closer, accommodation could not be ignored.
The measurement of eccentric acuity using this system may be useful in evaluating peripheral visual function of the parvocellular visual system. In the near future, this system will be used clinically for patients with parafoveal lesions and visual field defects.
References
1.Irving LD: Peripheral visual acuity. Am J Optom Physiol Optics 57:915-924, 1980
2.Green DG: Regional variations in the visual acuity for interference fringes on the retina. J Physiol 207:351-356, 1970
3.Mandelbaum J, Sloan LL: Peripheral visual acuity. Am J Ophthalmol 30:581-588, 1947
4.Loyd AJ, George CH: Photographic granularity and graininess. J Opt Soc Am 37:217-263, 1947
5.Millodot M: Foveal and extra-foveal acuity with and without stabilized retinal images. Br J Physiol Optics 23:75-106, 1966
6.Sloan LL: The photopic acuity-luminance function with special reference to parafoveal vision. Vision Res 8:901-911, 1968
7.Hart WM, Hartz RK, Hagen RW, Clark KW: Color contrast perimetry. Invest Ophthalmol Vis Sci 25:400-413, 1984
8 Frisen L: High-pass resolution targets in peripheral vision. Ophthalmology 94:1104-1108, 1987
9.Graham CH, Cook C: Visual acuity as a function of intensity and exposure time. Am J Psychol 49:654-661, 1937
10.Baron WS, Westheimer G: Visual acuity as a function of exposure duration. J Opt Soc Am 63:212219, 1973
11.Kono M, Yamade S: Temporal integration in diseased eyes. Int Ophthalmol 20:231-239, 1996
Title |
245 |
|
|
SHORT-WAVELENGTH AUTOMATED PERIMETRY IN NORMAL SUBJECTS
Prelimiary results*
DANIEL S. MOJON1 and MARIO ZULAUF2
1Department of Strabismus and Neuro-Ophthalmology, Kantonsspital, St Gallen; 2University Eye Clinic Basel, Basel; Switzerland
Abstract
Background: The purpose of this study was to evaluate short-wavelength automated perimetry (SWAP, i.e., blue-yellow) in normal volunteers and re-visit the normative values provided by the manufacturer. Methods: Twenty-eight eyes of 28 normal subjects (age range, 21-48 years; mean age, 36.5 years) were examined with SWAP (Octopus 101, two phases of program G2, Interzeag AG, Schlieren, Switzerland). All subjects had refractive errors with spherical equivalents < 5 diopters and astigmatism < 2 diopters, normal intraocular pressures, no history of diseases affecting the visual field or nerve fiber layer, and normal white-white automated perimetry (Octopus 101, program G2). Results: Twenty-one percent of the subjects (6/28) had to be excluded since visual field testing was not reliable (reliability factor > 5%). Forty-five percent of the remaining subjects (10/22) had all other indices within the instruments normative limits. With the appropriate normal values based on the multicenter SWAP Octopus 101 study, 11% (3/ 28) were beyond the normal range: all had abnormal high sensitivities – two due to false-positive replies. The normal-value range for the index mean defect (MD) is remarkably wide (5.-, median, 95th percentile, -4.4, -0.5, +5.3 dB, respectively). The normal-value range for the index loss variance (LV) is surprisingly low and similar to standard perimetry (5.-, median, 95th percentile, -1.7, 6.8, +21.2 dB2, respectively) . Conclusions: SWAP with the Octopus G2 program reaches an appropriate specificity, but only if the correct normal-values of the multicenter SWAP Octopus 101 study are used. The variability between subjects is remarkably large. The variability within a subject is similar for short-wavelength and standard perimetry, as reflected by the similar values for the visual-field index LV. Further studies are needed to establish the sensitivity of SWAP to detect a disease with the Octopus 101.
Acknowledgments
The authors have no financial interest in the products and instruments mentioned in this paper and have not received any financial reimbursement from Haag-Streit.
*A full version of this paper can be found in: Zulauf and Mojon: Normal values of short-wavelength automated perimetry. Ophthalmologica 217(4):260-264, 2003
Address for correspondence: Mario Zulauf, MD, University Eye Clinic Basel, CH-4012 Basel, Switzerland. Email: Mario.Zulauf@hin.ch
Perimetry Update 2002/2003, p. 245
Proceedings of the XVth International Perimetric Society Meeting, Stratford-upon-Avon, England, June 26–29, 2002
edited by David B. Henson and Michael Wall
© 2004 Kugler Publications, The Hague, The Netherlands