Measurement of iris color using computerized image analysis

Classification of Iris Colors and Patterns in Koreans

Objectives

The iris is the part of the eye that is colored, and it is an important feature in making an artificial eye. To address the lack of information about the brown Korean irises, we classified the colors and patterns of the irises of 100 Korean eyes.

Methods

The iris colors were extracted from 100 different digital photographs of each eye using K-mean clustering image analysis. The RGB data of the extracted colors were converted to CIELAB color space data and analyzed to determine the representative color of the iris in each photograph. The color differences were calculated to form groups of similar colors by matching the 100 samples into several groups with minimum color differences among them. The average hue value for each group was calculated, and the corresponding representative color was selected from 2 to 3 colors in one group.

Results

The iris colors were classified into seven brown color groups with 2 to 3 colors in each group. The coloration of Koreans' eyes appears to vary in a similar hue with differences in chroma and lightness. In addition, the iris patterns were classified into three basic patterns, namely, crown, sunshine, and flower, as well as three complex patterns that are mixtures of two basic patterns.

Conclusions

We classified Korean iris colors and patterns for application in the design of artificial eyes.

Tooth, skin, hair and eye colour interrelationships in Greek young adults

The purpose of this study was to investigate the possible interrelationships of teeth, skin, eye and hair colour. A portable colorimeter (Shade Eye NCC/Shofu) was used to record the colour in the CIELAB system of the upper right incisors in 150 dental school students, along with their skin colour at three different areas. Natural hair and eye colour was classified into several categories by a trained examiner (ICC 0.93-0.99). One-way ANOVA and correlation tests were used to statistically analyse the data. Skin was found to have significantly higher L*, b* but lower a* values than teeth (p < 0.05). A significant correlation (p < 0.05) of teeth to skin L* and a*colour coordinate was found, but not to b* coordinate (p > 0.05). Hair tones were not correlated to teeth L* or b*, but only to a*coordinate. Teeth and eye colour coordinates were not correlated (p > 0.05). Eye and hair tones were found to have the highest significant correlation (ρ = 0.369). In conclusion, teeth of this cohort were found to be lighter, less red and yellow than the skin. Teeth colour was not related to eye colour, but lighter teeth were found to be associated with lighter skins, and redder lateral incisors to lighter hair. Darker facial skins or yellower forehead areas were also associated with darker hair and vice versa. The clinical relevance of the study is that the investigated facial characteristics are inter-correlated weakly to moderately, and for this reason predicting the colour parameters of one facial characteristic by another would not be accurate, but helpful for a rough colour selection as associations show.

Genetics of human iris colour and patterns

The presence of melanin pigment within the iris is responsible for the visual impression of human eye colouration with complex patterns also evident in this tissue, including Fuchs' crypts, nevi, Wolfflin nodules and contraction furrows. The genetic basis underlying the determination and inheritance of these traits has been the subject of debate and research from the very beginning of quantitative trait studies in humans. Although segregation of blue-brown eye colour has been described using a simple Mendelian dominant-recessive gene model this is too simplistic, and a new molecular genetic perspective is needed to fully understand the biological complexities of this process as a polygenic trait. Nevertheless, it has been estimated that 74% of the variance in human eye colour can be explained by one interval on chromosome 15 that contains the OCA2 gene. Fine mapping of this region has identified a single base change rs12913832 T/C within intron 86 of the upstream HERC2 locus that explains almost all of this association with blue-brown eye colour. A model is presented whereby this SNP, serving as a target site for the SWI/SNF family member HLTF, acts as part of a highly evolutionary conserved regulatory element required for OCA2 gene activation through chromatin remodelling. Major candidate genes possibly effecting iris patterns are also discussed, including MITF and PAX6.

Increase and decrease of measurements of iris density of glaucoma patients treated with latanoprost over 36 months

PURPOSE

The aim of this study was to measure the density of iris photographs taken over time of patients treated with latanoprost.

METHODS

For 22 patients treated with latanoprost for 36 months, 250 iris photographs were obtained. Photographs were obtained at baseline, and then at 1 month on follow-up and subsequently at 4-month intervals. Photographs were measured in duplicate and masked for time, with a standardized method with red/blue light for density of color.

RESULTS

On visual inspection, 11 irides appeared to show an increase in color or pigment, while 12 showed no change. Eight (8) irides showed a significant, or borderline significant, increasing positive slopes of density measurements over time, while 9 irides showed a significant, or borderline significant, decreasing negative slopes. Six (6) subjects had no significant slope. There was a correspondence between the observations on visual inspection of an increase in color and the analysis of density measurements both by slope and by differences between final minus initial measurements. No leveling off of measurements was observed in almost all of the eyes, which showed increasing trends.

CONCLUSIONS

There is a heterogeneous response to treatment by latanoprost for 36 months in changes over time of red/blue density measurements for iris color.

Grading of Iris Color with an Extended Photographic Reference Set

Purpose

To present a new iris pigmentation classification system based on comparison of iris pigmentation to a set of 24 standard eye photographs, with the aim of gaining on accuracy and on applicability for retinal straylight studies.

Methods

A reference set of 24 eye photographs was established by ranking the photographs from least (number 1) to most (number 24) iris pigmentation. Reproducibility was tested by grading a sample of 67 eye photographs with this reference set.

Results

The overall variation between observers was 1.46 on a scale of 0 to 25.

Conclusion

The new method is promising to be more accurate than existing iris color classification systems in clinical situations where objective colorimetry-based systems are not available. The method may be useful to assess iris translucency and fundus reflectance as sources of variation in retinal straylight.

Digital analysis of mouthrinses? staining characteristics on provisional acrylic resins

Provisional restorations are expected to be both aestethically and physically durable during the preparation of permanent restorations. In this study, the staining properties of mouthrinses containing chlorhexidine gluconate, benzydamine hydrochloride and a hybrid mouthrinse were investigated on light and dark shades of a provisional acrylic resin. Totally 80 specimens were prepared and were photographed digitally to obtain the baseline L*, a*, b* values. Each sample was immersed in test solutions for 12 h which was equivalent time to 1 year of mouthrinse use, and the post-treatment images of the test materials were acquired. All L*, a*, b* values were analysed by a graphic software, and the total colour change (DeltaE*) of each specimen was calculated. Also the same colour analyses were performed on all test solutions to establish their colour parameters. Analysis of variance and Tukey's tests were used for statistical analyses and alpha was 0.05. All test solutions produced perceptible staining on the provisional material, with DeltaE values over 3.7. In both shades, hybrid rinse caused the highest staining (DeltaE=5.705), and was followed by chlorhexidine gluconate rinse, with DeltaE value of 4.120. The third highest staining was observed with benzydamine hydrochloride rinse (DeltaE=3.959), whereas the control caused the least staining (DeltaE=3.095). The lighter shade provisional material resulted with clinically observable staining even when immersed in distilled water; however, the dark shades showed clinically perceptible staining solely with the hybrid mouthrinse. In this study, the shade of the acrylic material was the determinator of the staining process.

Comparison of digital and spectrophotometric measurements of colour shade guides

Colour determination may be precisely performed with spectrophotometers, however this technique is time consuming and requires special devices. The aim of this study was to determine the correlation between the colour measurements performed on three shade guides of the same manufacturer, obtained with digital and spectrophotometric analyses. Sixty shade tabs of three shade guides were analysed to determine the L*, a*, b* values with a spectrophotometer. The same shade tabs were also imaged with a digital camera, and same colour parameters were determined on the digital images with the software. Three measurements for each shade tab were made with both methods. Statistical differences between the results of two methods were determined using anova (alpha = 0.05). Additionally, presence of any correlation between two methods within each guide was investigated by using correlation analysis. It was observed that the a* and b* values obtained by digital method were correlated with those of the spectrophotometer. Spectrophotometric analysis revealed no disparities among the L* measurements of the shade tabs of three shade guides (P > 0.05), but significant differences were observed between the L* values of the shade tabs with digital analysis method (P = 0.000). The digital method formed more subgroups than the spectrophotometer. For a* and b* values, both methods revealed significant differences among the shade tabs (P < 0.05). Additionally, both analyses revealed that colour characteristics of the first shade guide was different than the other two. The results obtained by digital method were correlated with those of the spectrophotometer, especially for a* and b* values.