Yu, Vincent Tak Cheong; (1992) A computer based method for determining defects of colour vision by measuring colour contrast thresholds in peripheral retina: Application to screening and management of populations at risk of developing glaucoma. Doctoral thesis (Ph.D), UCL (University College London).

Preview

Text A_computer_based_method_for_de.pdf Download (8MB) | Preview

Abstract

This thesis describes the details of the research and development of a new computer based technique for determining colour vision defects by measuring colour contrast thresholds in the peripheral region of the retina. The study shows that application of this new method to screening and management of populations at risk of developing glaucoma has great potential. The system consisted of a 80286 microprocessor based computer with a 80287 mathematics co-processor, a standard PC keyboard, a EGA monochromatic monitor; a computer graphic interface card containing an advanced CRT controller; and a high definition colour monitor. The visual stimulus was displayed on the high definition colour monitor and system information was shown on the EGA monitor. The system was controlled by a software written in Microsoft "C". A slit-lamp chin rest was utilised to position the patient's head. A closed circuit infra-red sensitive camera with a video monitor was used to monitor the eye movements of patients. The visual stimulus was an annulus displayed on a background which was equiluminant but of a contrasting colour. Equiluminance was achieved for each subject by obtaining a luminance match between the red and green guns and blue and green guns of the TV monitor, using heterochromatic flicker photometry. The annulus subtended 25 degrees at the eye, and 45 degrees of the annulus was removed randomly in one of the four quadrants. The colour contrast between the annulus and the background could be varied along protan, deuteran or tritan colour confusion lines, and the test consisted of establishing the minimum colour-contrast between annulus and background which allowed patients to identify the position of the gap while regarding a central fixation spot. Thus, the test was a "4-way forced choice" psychophysical method. A modified binary search method was used to ensure that the stimuli converged rapidly toward colour contrast threshold. Tests were done monocularly. The eye not being tested was occluded. A series of control experiments were carried out to test the reliability and limitations of the system which included investigation of the effect of age, refractive error, eccentricity, pupil size, and illumination level. The linear regression of the test-retest results is 0.984. All patients tested so far preferred this new test to automated perimetry. 325 eyes had been tested which included 84 controls, 84 glaucoma, 77 high risk, 50 medium risk and 30 low risk ocular hypertensives. None of the glaucoma results overlapped those of the controls: all glaucoma thresholds were higher than 3.A SDs above the normal mean. Over 50 % of the high risk thresholds were higher than 3 SDs above normal mean; some medium risk and a few low risk thresholds had high threshold values as well. The spread of thresholds between normal and glaucomatous patients is considerable, so many intermediate graduations could be recognised. Thus, for screening purposes, criteria can be set such that a desired proportion of each class of patients may be separated from the general population. The high reliability and sensitivity of discriminating between glaucoma patients and normals suggests that this test can be proved of great value in screening glaucoma and early detection of persons in the population at risk of developing glaucoma.

Download activity - last month

Export as JSONCSVXML

Download activity - last 12 months

Export as JSONCSVXML

Downloads by country - last 12 months

Export as JSONCSVXML

Archive Staff Only

View Item