Colour Assessment And Diagnosis Research Articles

Assessment of novel binocular colour, motion and contrast tests in glaucoma

The effects of glaucoma on binocular visual sensitivity for the detection of various stimulus attributes are investigated at the fovea and in four paracentral retinal regions. The study employed a number of visual stimuli designed to isolate the processing of various stimulus attributes. We measured absolute contrast detection thresholds and functional contrast sensitivity by using Landolt ring stimuli. This psychophysical Landolt C-based contrast test of detection and gap discrimination allowed us to test parafoveally at 6 ° from fixation and foveally by employing interleaved testing locations. First-order motion perception was examined by using moving stimuli embedded in static luminance contrast noise. Red/green (RG) and yellow/blue (YB) colour thresholds were measured with the Colour Assessment and Diagnosis (CAD) test, which utilises random dynamic luminance contrast noise (± 45 %) to ensure that only colour and not luminance signals are available for target detection. Subjects were normal controls (n = 65) and glaucoma patients with binocular visual field defects (n = 15) classified based on their Humphrey Field Analyzer mean deviation (MD) scores. The impairment of visual function varied depending on the stimulus attribute and location tested. Progression of loss was noted for all tests as the degree of glaucoma increased. For subjects with mild glaucoma (MD -0.01 dB to -6.00 dB) significantly more data points fell outside the normal age-representative range for RG colour thresholds than for any other visual test, followed by motion thresholds. This was particularly the case for the parafoveal data compared with the foveal data. Thus, a multifaceted measure of binocular visual performance, incorporating RG colour and motion test at multiple locations, might provide a better index for comparison with quality of life measures in glaucoma.

Colour vision as a screening tool for sub‐clinical retinal disease

AbstractPurpose Age‐related retinal diseases start developing at a subclinical level, as gradual photoreceptor dysfunction, ganglion cell apoptosis etc. Subtle retinal changes disturb the fine intraretinal signal transmission and despite the lack of clinically detectable signs can cause an impairment of chromatic sensitivity that precedes the loss of performance in processing other stimulus attributes. Colour vision can be used as a sensitive screening tool for retinal diseases when the tests are carried out at lower light levels.Methods The health of the retina (HR) index was introduced to detect early changes caused by disease by capturing the worsening of chromatic sensitivity as the light level changes from photopic to mesopic (J Opt Soc Am A Opt Image Sci Vis 29,2:A27‐A35,2012). This approach makes it possible to separate changes caused by normal aging from the effects of sub‐clinical retinal disease. Chromatic sensitivity was measured in 60 subjects (mean age 47.6 years, range 16‐79 years) at four light levels using the Colour Assessment and Diagnosis (CAD) test; measurement of pre‐receptoral absorption by the crystalline lens and the macular pigment and the size of the pupil allowed an estimation of retinal illuminance.Results The HR index proved largely independent of age (r2~0.2), but ~11% of clinically normal older subjects showed impaired chromatic sensitivity that could reflect the development of retinal pathology. Some of these subjects developed AMD and Glaucoma 2.5 years after the assessment. Patients with identified retinal disease also show abnormal HR indices.Conclusion The HR index is a single number that captures information about the health of the retina irrespective of age and can be used to screen for sub‐clinical signs of retinal diseases.

Age‐corrected monocular limits of normal trichromacy

AbstractPurpose Chromatic sensitivity worsens gradually with increasing age, but the inherent variability in normal colour vision makes this difficult to detect and to separate from very early changes caused by either retinal or systemic diseases. In this study we developed new approaches to identify and screen for changes that cannot be attributed to normal aging. This approach made it possible to produce reliable, age‐corrected limits for normal colour vision.Methods Red‐green (RG) and yellow‐blue (YB) thresholds were measured monocularly in each eye (using the Colour Assessment and Diagnosis (CAD) test) in 394 subjects (age range: 16 to 90 years). Only clinically normal subjects according to medical history and eye / fundus examination were included. All subjects with congenital colour deficiency were excluded. A new filter based on right / left eye asymmetry for RG and YB thresholds was also employed to exclude subjects with significant differences in monocular thresholds. After filtering, the analysis of the effects of normal aging was based on 178 subjects (age range: 16‐79 years).Results The effects of age (in the range 20 to 80 yrs) can be accounted for well by a linear trend with just over two‐fold increase in both RG and YB mean colour thresholds. Statistical limits of variability (±2σ) were computed from the normal subject data in order to provide limits of age‐corrected RG and YB colour vision.Conclusion The age‐corrected CAD thresholds limits established in this study provide a more accurate way of detecting monocular changes in colour vision caused by either retinal diseases or systemic conditions that affect the eye.

Detection of colour vision changes in patients with systemic diseases that can affect the eye

AbstractPurpose Changes in colour vision can provide the earliest signs of vision loss caused by either retinal or systemic disease (Expert Rev.Ophthalmol. 6(4):409‐420,2011). Both yellow‐blue (YB) and red‐green (RG) mechanisms are affected in acquired deficiency, but some diseases may cause different patterns of colour vision loss. Patients with different systemic diseases were examined to discover abnormal changes in colour vision using a newly developed, age‐corrected system for colour assessment.Methods 334 subjects (age range: 41 to 91years) were investigated. The subjects were classified as normal, hypertensive, diabetic, diabetic and hypertensive, idiopathic and other systemic conditions. RG and YB colour thresholds were measured monocularly in each eye using the Colour Assessment and Diagnosis (CAD) test.Results Subjects with early stage pathology were excluded using a number of criteria including the discovery of right / left eye asymmetric loss in colour sensitivity. Age‐corrected limits of normal colour vision were employed to detect significant changes in chromatic sensitivity. The results reveal significantly higher RG and YB thresholds in all groups (including the idiopathic and other systemic conditions group).Conclusion The use of age‐corrected normal limits of chromatic sensitivity makes it possible to detect reliably significant loss of colour vision in systemic disease. Loss of chromatic sensitivity that precedes retinopathy is present in every systemic condition examined. Further studies are needed to discover the extent to which the pattern of selective RG and YB losses can be attributed to specific diseases of the eye.

Processing of color signals in female carriers of color vision deficiency

The aim of this study was to assess the chromatic sensitivity of carriers of color deficiency, specifically in relation to dependence on retinal illuminance, and to reference these findings to the corresponding red-green (RG) thresholds measured in normal trichromatic males. Thirty-six carriers of congenital RG color deficiency and 26 normal trichromatic males participated in the study. The retinal illuminance was estimated by measuring the pupil diameter and the optical density of the lens and the macular pigment. Each subject's color vision was examined using the Color Assessment and Diagnosis (CAD) test, the Ishihara and American Optical pseudoisochromatic plates, and the Nagel anomaloscope. Carriers of deuteranopia (D) and deuteranomaly (DA) had higher RG thresholds than male trichromats (p < 0.05). When referenced to male trichromats, carriers of protanomaly (PA) needed 28% less color signal strength; carriers of D required ∼60% higher thresholds at mesopic light levels. Variation in the L:M ratio and hence the absolute M-cone density may be the principal factor underlying the poorer chromatic sensitivity of D carriers in the low photopic range. The increased sensitivity of PA carriers at lower light levels is consistent with the pooling of signals from the hybrid M' and the M cones and the subsequent stronger inhibition of the rods. The findings suggest that signals from hybrid photopigments may pool preferentially with the spectrally closest "normal" pigments.

Hypoxia, color vision deficiencies, and blood oxygen saturation

Chromatic thresholds were measured using the Cambridge Colour Test (CCT), the Colour Assessment and Diagnosis (CAD) test, and the Cone Specific Contrast Test (CSCT) at ground and 3780 m (12,400 ft) for subjects with normal color vision and red-green color vision defects. The CAD revealed a small (~10%) increase in the red-green thresholds for the trichromatic subjects and a similar increase in the blue-yellow thresholds for the dichromats. The other two color vision tests did not reveal any significant change in chromatic thresholds. The CAD results for the trichromats were consistent with a rotation of the discrimination ellipse counterclockwise with little change in the elliptical area. This alteration in the color discrimination ellipse can occur when retinal illumination is lowered.

Assessing the Severity of Color Vision Loss with Implications for Aviation and Other Occupational Environments

The Ishihara Test (IT) is arguably the most sensitive and commonly used color vision test within aviation and other occupational environments, but when no errors are allowed -20% of normal trichromats fail the test. The number of allowed errors varies in different occupations and sometimes within the same environment (such as aviation) in order to reflect the difficulties of the color-related tasks. The implicit assumption is that the plates can be ranked in order of difficulty. The principal aim of this study was to investigate whether appropriate "weights" can be attached to each IT plate to reflect the likelihood of producing a correct response. A second aim was to justify the use of color thresholds for quantifying the loss of red-green (RG) and yellow-blue (YB) chromatic sensitivity. We investigated 742 subjects (236 normals, 340 deutans, and 166 protans) using the first 25 plates of the 38-plate IT and measured RG chromatic sensitivity using the Color Assessment and Diagnosis (CAD) test. The IT error scores provided plate-specific "weights" which were used to calculate a Severity Index (SI) of color vision loss for each subject. Error scores, SI values, and CAD thresholds were measured and compared in each of the three subject groups. Color thresholds can provide a good measure of the severity of both RG and YB color vision loss. Neither the number of IT plates failed nor the SI value computed in this way can be used to determine reliably the severity of color vision loss.

Oxygenation State and Twilight Vision at 2438 m

Under twilight viewing conditions, hypoxia, equivalent to breathing air at 3048 m (10,000 ft), compromises low contrast acuity, dynamic contrast sensitivity, and chromatic sensitivity. Selected past experiments have been repeated under milder hypoxia, equivalent to altitude exposure below 2438 m (8000 ft), to further define the influence of oxygenation state on mesopic vision. To assess photopic and mesopic visual function, 12 subjects each undertook three experiments using the Contrast Acuity Assessment test, the Frequency Doubling Perimeter, and the Color Assessment and Diagnosis (CAD) test. Experiments were conducted near sea level breathing 15.2% oxygen (balance nitrogen) and 100% oxygen, representing mild hypobaric hypoxia at 2438 m (8000 ft) and the benefit of supplementary oxygen, respectively. Oxygenation state was a statistically significant determinant of visual performance on all three visual parameters at mesopic, but not photopic, luminance. Mesopic sensitivity was greater with supplementary oxygen, but the magnitude of each hypoxic decrement was slight. Hypoxia elevated mesopic contrast acuity thresholds by approximately 4%; decreased mesopic dynamic contrast sensitivity by approximately 2 dB; and extended mean color ellipse axis length by approximately one CAD unit at mesopic luminance (that is, hypoxia decreased chromatic sensitivity). The results indicate that twilight vision may be susceptible to conditions of altered oxygenation at upper-to-mid mesopic luminance with relevance to contemporary night flying, including using night vision devices. Supplementary oxygen should be considered when optimal visual performance is mission-critical during flight above 2438 m (8000 ft) in dim light.

Colour detection thresholds as a function of chromatic adaptation and light level

Colour threshold discrimination ellipses were measured for a number of states of chromatic adaptation and a range of luminance levels using the Colour Assessment and Diagnosis (CAD) test. An analysis of these results was carried out by examining the cone excitation signals along the cardinal axes that correspond to detection thresholds in the +L-M (reddish), -L+M (greenish), +S (bluish) and -S (yellowish) colour directions. The results reveal a strong linear relationship between the excitations induced by the adapting background field in each cone class and the corresponding changes needed for threshold detection. These findings suggest that the cone excitation change for threshold detection of colour signals is always the same for a given background excitation level (in any cone class), independent of the excitations generated in the other cone classes. These observations have been used to develop a model to predict colour detection thresholds for any specified background luminance and chromaticity within the range of values investigated in this study (e.g., luminances in the range 0.3 to 31 cd.m(-2) and chromaticities within the gamut of typical CRT displays). Predicted colour thresholds were found to be in close agreement with measured values with errors that do not, in general, exceed the measured within-subject variability.

Loss of chromatic sensitivity in AMD and diabetes: a comparative study

Abnormalities in rod and cone photoreceptor morphology have been reported in normal aging retinas in the absence of known pathology and have been taken as an indicator of susceptibility to retinal disease. Some loss of visual performance may therefore precede retinal structural changes that can be detected reliably using conventional fundus imaging techniques. Red/green (RG) and yellow/blue (YB) colour discrimination thresholds are sensitive measures of normal retinal function and poor YB discrimination is often taken as an indicator of retinal disease, though it is generally acknowledged that RG loss is also present in most cases of acquired deficiency. Although structural changes in age-related macular degeneration (AMD) and diabetes share some similarities, significant differences remain and this may result in different patterns of RG and YB loss. The aim of this study was to quantify and compare the severity of RG and YB loss of chromatic sensitivity in patients with AMD and diabetes. Patients with varying severity of AMD and diabetes and normal subjects of similar age were recruited for the study. RG and YB colour detection thresholds were measured in the two groups of patients and the control group, using the Colour Assessment and Diagnosis (CAD) test. Each AMD subject investigated showed significant, but unequal loss of YB and RG chromatic sensitivity, with YB discrimination showing the greatest loss. Diabetic subjects also exhibited reduced chromatic sensitivity, but with almost equal and highly correlated RG and YB thresholds (R(2) = 0.99). The severity of colour vision loss measured with the CAD test also correlates well with a clinical classification index of disease progression in AMD, and with the level of hyperglycaemic control in diabetes. These findings suggest that accurate measurements of RG and YB colour thresholds can provide a sensitive measure of functional change in diseases of the retina with patterns of loss that differ significantly in AMD and diabetes.

Differential effects of ageing on foveal and peripheral colour vision

Abstract Purpose Colour sensitivity was assessed to establish aging effects both at the fovea and 6 deg away from fixation, in each of the four quadrants. Methods 65 normal healthy subjects (from 20 to 80 years of age) took part in the study. All subjects had Visual Acuity (VA) of 6/6 or better. Fixation accuracy was monitored using infrared imaging of the pupil and the tests were carried out on the P_SCAN system. Target size was adjusted for parafoveal locations to account for retinal and cortical magnification. Yellow‐blue (YB) and red‐green (RG) colour discrimination was assessed using the CAD (colour assessment and diagnosis) test (http://www.caa.co.uk/docs/33/200904.pdf). Results RG and YB colour thresholds were analysed separately for all five locations tested and showed no significant effect with ageing below the age of 60 years. Two age bands were formed based on statistical analysis (20‐59.9 and 60‐79.9). The decline in performance with age was more rapid at the fovea and exhibited a steeper gradient when compared with results in the periphery for both RG and YB discrimination. Foveal YB discrimination showed the largest ageing effect. No significant difference was found between the four parafoveal locations. YB discrimination at the fovea also exhibited the largest inter‐subject variability. Conclusion These findings may have clinical significance in the very early detection of disease processes that remain subclinical in many subjects. Differences between foveal and peripheral locations help to differentiate between the normal effects of ageing and disease. For example, higher foveal and normal peripheral YB thresholds in normal subjects from high peripheral thresholds in early glaucomatous subjects.

Pre‐receptoral spectral absorption, healthy ageing and pre‐clinical indications of retinal disease

Abstract Purpose The aim of this study was to investigate how chromatic sensitivity changes as a function of age and to establish the extent to which such changes can be attributed to pre‐receptoral spectral absorption of short wavelength light and/or changes in retinal mechanisms caused by ageing. Methods The absorption of blue light by the macular pigment (MP) and the crystalline lens and the subjects’ sensitivity to rapid flicker were measured using the Macula Assessment Profile (MAP) test. Red‐green (RG) and yellow blue (YB) chromatic detection thresholds were measured at the fovea for young and older subjects using the Colour Assessment and Diagnosis (CAD) test at 2.6, 26 and 65 cd/m2. The variables of interest included the spectral absorption of the lens, the optical density of the MP, subject’s age and retinal illuminance. Results The absorption of blue light by the lens increased with age. Absorption of blue light by pre‐receptoral filters did not affect RG chromatic sensitivity at any of the light levels investigated but had an effect on YB thresholds. The considerably higher colour vision thresholds of some subjects and the subsequent worsening of their chromatic sensitivity at the lower light level may reflect changes in the retina brought about by accelerated aging effects. Conclusion The effect of pre‐receptoral absorption of blue light on chromatic sensitivity is small. Ageing affects the amount and spectral composition of the light reaching the photoreceptors and the processing of retinal signals. As a result, flicker sensitivity declines and colour vision deteriorates. Such effects arise mostly from changes in the retina. The MAP and CAD tests help us to detect the effects of accelerated ageing and retinal disease.

Sex-related differences in chromatic sensitivity

Generally women are believed to be more discriminating than men in the use of color names and this is often taken to imply superior color vision. However, if both X-chromosome linked color deficient males (8%) and females (<1%) as well as heterozygote female carriers (15%) are excluded from comparisons, then differences between men and women in red-green (RG) color discrimination have been reported as not being significant (e.g., Pickford, 1944; Hood et al., 2006). We re-examined this question by assessing the performance of 150 males and 150 females on the color assessment and diagnosis (CAD) test (Rodriguez-Carmona et al., 2005). This is a sensitive test that yields small color detection thresholds. The test employs direction-specific, moving, chromatic stimuli embedded in a background of random, dynamic, luminance contrast noise. A four-alternative, forced-choice procedure is employed to measure the subject's thresholds for detection of color signals in 16 directions in color space, while ensuring that the subject cannot make use of any residual luminance contrast signals. In addition, we measured the Rayleigh anomaloscope matches in a subgroup of 111 males and 114 females. All the age-matched males (30.8 +/- 9.7) and females (26.7 +/- 8.8) had normal color vision as diagnosed by a battery of conventional color vision tests. Females with known color deficient relatives were excluded from the study. Comparisons between the male and female groups revealed no significant differences in anomaloscope midpoints (p = 0.709), but a significant difference in matching ranges (p = 0.040); females on average tended to have a larger mean range (4.11) than males (3.75). Females also had significantly higher CAD thresholds than males along the RG (p = 0.0004), but not along the yellow-blue (YB) discrimination axis. The differences between males and females in RG discrimination may be related to the heterozygosity in X-linked cone photo pigment expression common among females.

Mild Hypoxia Impairs Chromatic Sensitivity in the Mesopic Range

The effect of mild hypoxia on chromatic sensitivity in the mesopic range is poorly documented. This study was conducted to examine the effects of mild hypoxia and hyperoxia on red-green (R-G) and yellow-blue (Y-B) chromatic sensitivity thresholds at low photopic (22.3 cd . m(-2)), borderline upper mesopic (1.67 cd . m(-2)) and mid-mesopic (0.21 cd . m(-2)) luminance. The Color Assessment and Diagnosis (CAD) test was used to measure binocular and monocular R-G and Y-B chromatic sensitivity by using dynamic luminance contrast noise to isolate the use of color signals. Mild hypoxia was imposed by breathing 14.1% oxygen and was investigated relative to control exposures breathing air (normoxia) at each light level. Subsequently, hyperoxia, breathing 100% oxygen, was assessed relative to hypoxia under the mesopic conditions. A balanced, repeated-measures design allowed assessment of main effects and interactions of light level, viewing condition, gender, breathing gas, and exposure order by using multivariate analysis of variance (MANOVA), with post hoc analysis employing ANOVA and paired t-tests. Light level, number of viewing eyes, and oxygenation state were significant determinants of chromatic sensitivity. One man and one woman introduced orthogonal sources of gender bias. The CAD test revealed minimal deuteranomaly (R-G deficiency) in the man and loss of Y-B sensitivity in the only woman using hormonal contraception. In the mesopic range, mild hypoxia impairs chromatic sensitivity progressively with reducing luminance. Binocular summation of chromatic signals is consistent and independent of the luminance channel. The CAD test is highly sensitive to mild congenital and acquired color vision deficiencies.

Recovery of vision and pupil responses in optic neuritis and multiple sclerosis

The recovery of visual performance and pupil responses were investigated in patients with demyelinating optic neuritis (ON) and multiple sclerosis (MS). The pupil constriction amplitude and the time delay (latency) of the pupil response were measured in 14 patients with a history of unilateral ON in response to either achromatic (luminance) or chromatic (isoluminant) stimulus modulation. Five of these subjects were diagnosed later with MS. In addition, we measured detection thresholds for achromatic stimuli using standard visual field perimetry and chromatic thresholds using a new colour assessment and diagnosis (CAD) test that isolates the use of colour signals. The results show that, despite significant improvements in visual function following the acute phase (as assessed using visual acuity and fields), significant pupil response deficits remain. The findings also demonstrate that accurate measurements of pupil responses and chromatic thresholds can reveal deficits that remain undetected with more conventional techniques. These preliminary findings suggest that the techniques described here can provide useful information about remitting and relapsing demyelinative phases, often observed during MS and ON.

Evaluation of the New Web-Based “Colour Assessment and Diagnosis” Test

The purpose of the study was to determine the sensitivity, specificity, and repeatability of the web-based Colour Assessment and Diagnosis (CAD) test in comparison to current tests of color vision. Thirty color normals and 30 color deficients, identified and diagnosed by the Nagel anomaloscope, were tested. The results of the CAD test were compared with standard tests like Nagel anomaloscope, Ishihara (concise version, 2001), Hardy, Rand and Rittler (HRR; 4 ed) pseudoisochromatic test, and the Farnsworth Munsell 100 (FM-100) hue test. Using the Nagel anomaloscope as the "gold standard," the sensitivity with the CAD test was 93.33%, Ishihara 96%, HRR 100%, and the FM-100 hue 100%. The specificity was 100% with CAD and the Ishihara color plates, whereas it was 33% with the HRR and 83% with the FM-100 hue test. The concurrent validity of the CAD test for color normals was 93.75%. The concurrent validity of CAD test for color deficiency was 100%. Thus, anyone failing the CAD test has a color defect. The coefficient of agreement for the Nagel anomaloscope and the CAD test was 0.93, with Ishihara it was 0.96, with the HRR it was 0.33, and with FM-100 hue it was 0.83. These results showed that the CAD test is a valid test for identifying congenital red-green color deficiency. Further testing is required in a larger population of anomalous trichromats.