Premium
Grading iris color of post‐mortem human eyes
Author(s) -
Madigan M.,
Cionaca V.,
Sitiwin E.,
Ton H.T.
Publication year - 2016
Publication title -
acta ophthalmologica
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.534
H-Index - 87
eISSN - 1755-3768
pISSN - 1755-375X
DOI - 10.1111/j.1755-3768.2016.0670
Subject(s) - iris (biosensor) , hue , grading scale , grading (engineering) , artificial intelligence , human eye , brightness , color space , iris recognition , computer vision , medicine , computer science , optics , biometrics , physics , biology , surgery , ecology , image (mathematics)
Purpose Iris colour encompasses a continuum from pale blue to very dark brown and is usually classified via a descriptive three colour scale: blue, green‐hazel or brown. Digital imaging technologies provide an opportunity to quantify iris colour, and are increasingly used for studies of genetic variations in iris colour. We explored the use of digital imaging and colour space information for grading of iris colour in post‐mortem eyes, including eyes with choroidal naevi. Methods Post‐mortem adult human irises ( n = 25) were examined and photographed using a Jenoptik digital camera and ProgResCaptureProv2.8.9 software. Standard lighting (colour temperature) and parameters for imaging were established and used for all specimens. Iris colour was graded ( n = 5) (using a nine‐category system; Mackey et al. Clin Exp Ophthalmol, 2011). We also developed a method in Photoshop to express iris colour as Hue in the Hue, Saturation, Brightness (HSB) colour space, for comparison with the category grading. Results Using the nine‐category grading end‐point grades (light blue and dark brown) were consistently graded. Intermediate grades were categorised differently for some observers, usually with adjacent categories. Green irises were not observed in this small series. Digital imaging using standard iris images and an averaging filter, provided colour information (Hue) for each iris. This allowed discrimination of iris colour compared to category grading. Conclusions As expected category‐grading was not always consistent between observers. We developed a digital imaging approach using HSB color space to give a H value for each iris. We are exploring the utility of processing functions such as gaussian blur. This approach provides a standard iris colour for post‐mortem tissue and will be used for comparison with fundus colour.