Age-related Macular Degeneration (AMD) is a degenerative disorder that accounts for about 50% of blindness in England and Wales. At present there is no effective treatment. It occurs in genetically susceptible individuals exposed to environmental factors such as smoking but so far specific factors remain to be identified. For this descriptive study, 879 patients with Age-Related Maculopathy (ARM) and AMD and 44 spouses with normal maculae, to act as a comparison group, were recruited from a tertiary referral centre. The clinical phenotypes were analysed from fundus photographs, fluorescein angiography and autofluorescence (AF) images. Fundus features were characterised as they are thought to reflect the genes conferring risk in an individual and may allow greater understanding of disease mechanisms. These data demonstrated (1) A revised grading system shown to be reproducible for use with digital images (2) A moderate concordance rate for phenotype between eyes with end- stage AMD (kappa statistic=0.48 95% CI = 0.38-0.57, p0.001). (3) Distinct characteristics, including a larger area and higher counts of soft drusen with focal areas of increased AF, in fellow eyes of those with unilateral visual loss due to geographic atrophy (GA). (4) An increased susceptibility of the inferotemporal macula to GA. (5) Preserved integrity of the retinal pigment epithelium (RPE) in the initial stages of CNV development as identified from AF images. (6) Loss of scotopic rather than photopic function over areas of increased AF, as determined by fine matrix mapping, indicating the preferential vulnerability of rods. (7) No difference in smoking history between those with neovascular compared to non-neovascular AMD. Although AMD has been extensively investigated, this study extends our knowledge of retinal AF, the relative susceptibility of rods compared to cones at the macula and suggests both eyes of an individual are more discordant for late stages of AMD compared to drusen.

Colour is used extensively in the railway industry to enhance conspicuity, code information and group objects of interest together. An acceptable level of colour discrimination is therefore required to enhance visibility even when colour is used redundantly, thus justifying the need to adequately screen all employed personnel. Rail Safety and Standards Board (2002, 2007) state that colour vision must be normal, as assessed by the Ishihara Plates Test or City University test. However, current conventional screening tests are limited in design and have high variability. As a result, a study was commissioned by Transport for London (TfL) to derive empirically, new limits of colour vision loss that could be classed as safe for train operators, within the London Underground (LU) environment. An initial visual task analysis was conducted to indentify the most safety-critical, colour-related task that LU train drivers encountered. This was determined to be the 'within tunnel' signal lights. Unique to this study was the original and faithful laboratory construction and reproduction of the within-tunnel signal light task which made it possible to establish experimentally new colour limits. A total of 100 subjects were recruited for the study, 40 normal trichromats and 60 congenital colour deficient'. They were assessed on conventional screening tests, the Colour Assessment and Diagnosis (CAD) test and compared against the simulated Trains Light (TL) test. Safe colour vision limits were established for two approach distances of 220 m and 110 m. Analysis indicated that for an approach distance of 110 m, deuteranomalous subjects with RG CAD thresholds < 7 and protanomalous subjects with RG CAD thresholds < 10 performed the TL test as well as normal trichromats and were therefore classed as safe to operate a train. An extended analysis of the proposed limits which included previous study data and LU data was conducted. A total of 606 subjects' data was reviewed (normal n = 205, deutan n = 269, protan n = 132) to provide accurate predictive outcomes. Based on the proposed 110 m limits, - 40% of colour deficient applicants would be classed as safe to drive a train. If the approach distance of 220 m were adopted, - 11 % of colour deficient applicants would be classed as safe to conduct a train based on stricter RG CAD thresholds limits of < 2.5 for deuteranomalous subjects and < 9 for protanomalous subjects. TfL have now accepted the CAD limits proposed for an approach distance of 110 m, classifying - 40% of the colour deficient population as safe. The CAD system now replaces the Ishihara test in the LU occupational health centres, and is used during initial and renewal certification of medical fitness assessments. The second study focused on acquired visual function loss associated with ageing and retinal disease. Studies had suggested that visual acuity measurements were not a good predictor of early retinal disease where a significant loss in VA was perceptible only if the majority of the photoreceptors were dysfunctional (Sunness et al., 1997; Gellar et al., 1992). Coupled with the high inter-session measurement variablitiyfor visual acuity (Patel et al., 2008), the aim of the study was to identify the most appropriate and sensitive clinical test capable of detecting early morphological changes associated with early Age-related Macular Degeneration (AMD). A total of 71 subjects were recruited for the study, a normal control group (n = 45) and an active AMD group (n = 26) of varying disease severity. Subjects underwent extensive clinical screening to ensure they adhered to strict,inclusion criteria. Advanced psychophysical assessment of visual function for visual acuity, contrast sensitivity, mesopic and photopic temporal and chromatic sensitivity were conducted. Exclusive to this study was the refinement of a new temporal resolution test. Significant clinical predictors for early-AMD were determined based on bivariate regression analysis and entered stepwise into a multiple linear regression model, controlling for eye and age. Yellow-blue (YB) chromatic discrimination was identified as the best clinical predictor, explaining an additional 9 %-12 % of the variance for early-AMD detection and - 22 %-25 % for AMD disease. Receiver operating characteristic (ROC) and area under the curve (AUC) analysis confirmed YB chromatic sensitivity to be 13.5 % more sensitive and 16.7 % more specific than the standard clinical Log MAR acuity test for detecting early-AMD retinal changes. A review of AMD and diabetic case studies also indicated that accurate Psychophysical methods of assessing chromatic sensitivity enhanced our clinical ability to detect, quantify and monitor changes associated with retinal disease and / or treatment outcomes.

This thesis presents an automated framework for quantitative analysis of fluorescein angiographies of the human retina. Such framework represents the core of a computer-aided system, which can assist NHS clinicians in early diagnosis of macular degeneration (AMD). The presented methodology aims to demonstrate the technical feasibility of automated extraction of retinal blood flow parameters, and results in a step forward in the development of an automated computer vision system for quantitative analysis of fluorescein angiograms to assist NHS clinicians in early diagnosis of AMD. The approach commences by segmenting the anatomic constituent parts of the ocular fundus, i.e. the optic disc (OD), the fovea and the vascular network tree. The OD/fovea are simultaneously detected by combining luminance information and geometric information from the major blood vessels; information regarding OD/fovea is then used for delineating the macula. Meanwhile, to segment the retinal vasculature, three independent approaches are implemented. These approaches use information about maximum curvature in both image- and frequency-domains. Such information is combined, firstly, using a supervised linear classifier. Secondly, utilising a committee of local experts, where each expert is represented by an artificial neural network. A fuzzy clustering algorithm is used for expert selection based on the specific input pattern. The output of the system is determined through a winner-takes-all rule. And finally. by considering a tracing algorithm that follows vessel centrelines and walls using a set of rules based upon information of maximum curvature and symmetry. Following segmentation, the extracted vasculature is utilised as input features for a multi modal registration algorithm, which has its fundamentals on the Fourier transform and a parametric estimation based on the gradient of the quadratic error function and least squares computation. Once subsequent frames of the angiogram have been aligned, anatomic, morphologic and sequential analyses are carried out. Special attention is given to the latter one, which is a methodology for quantitative analysis of retinal haemodynamics. It analyses retinal blood flow based on the estimation of parameters such as mean transit-time (MTT) and vascular volume. The former parameter is estimated using densitometry and analysis of the vascular response; the latter is calculated from the lumen of extracted vessels. The performance of the framework is demonstrated on a comprehensive dataset, which contains images of normal retinas and retinas with pathologies such as wet age-related macular degeneration and branch retinal vein occlusion. Results achieved in certain individual modules overcame serious defects observed in previous methods.

Retinitis Pigmentosa (RP, MIM#268000) is a significant cause of blindness in the Western world with an incidence of 1 in 3500. The mode of inheritance in RP may be autosomal recessive (ar), autosomal dominant (ad), X-linked recessive or digenic. To date, 39 loci have been implicated in RP, of which 30 genes are known (http://www.sph.uth.tmc.edu/Retnet/home.htm). Three recently identified pre-mRNA splicing factors, PRPF31, PRPF8, and PRPF3 correspond to RP11, 13, and 18, respectively. Our ongoing research on the genetic basis of RP resulted in the mapping of an autosomal dominant RP (adRP) locus on chromosome 19q13.4 (RP11). The genetic interval was refined to a 520kb region flanked by markers D19S927 and D19S781.2 Mutation screening led to the exciting discovery of splicing factor gene, PRPF31, implicated in RP11. The unique feature associated with this locus has been the consistent finding of partial penetrance phenotype in most RP11 families worldwide. The molecular mechanism responsible for this feature still remains unclear. We speculated that the reason for partial penetrance is due to the different levels of expression of the normal copy of the gene that is able to compensate for that mutant allele is asymptomatic individuals (or unable to in the symptomatic individuals). Partial penetrance can be viewed as nature’s way of curing the disease. Though our recent studies we demonstrated that symptomatic and asymptomatic individuals in a given sib-ship consistently inherited different wild-type (WT) alleles of pRPF from their unaffected parent. This resulted in the asymptomatics having higher levels of WT PRPF31 mRNA compared to symptomatic individuals. We feel this might explain the partial penetrance phenotype observed in most RP11 families. The aim of the thesis is to explore the molecular mechanism that underlies the partial penetrance phenotype associated with the PRPF31 gene, causing adRP. For this reason, the variation in the wild-type PRPF31 expression level was assessed in random unrelated individuals, at the mRNA level and protein level. From the data generated, it showed that in the general population there are individuals with two high expressing alleles (HH), individuals with one high and one low (HL) and others who have two low expressing alleles (LL), suggesting a continuous spectrum with three overlapping (high, intermediate and low) levels of PRPF31 expression. In order to identify the novel sequence change in PRPF31 gene, responsible for the differential expression of PRPF31 alleles, the ~520kb region spanning PRPF31 gene has been targeted to be amplified and sequenced, where the genetic change might exist. A SNP rs460824 located #93 kb (based on the standard ENSEMBL database) upstream of PRPF31 showed significant association (chi square 7.5, P<0.01) in a RP11 family. This particular SNP analysis highlighted the fact that there could be other SNPs across the 520kb interval which may act in combination to up-regulate or down-regulate the transcriptional activity. In addition, the presence of copy number variation at the RP11 locus was also investigated to identify as a cause for the differential expression of PRPF31. Comparative genomic hybridization (CGH) tiling array and real time quantitative PCR study revealed a difference in copy number downstream of PRPF31 gene on chromosome 19q13.4 between coordinates 59410000-59450000, in individuals homozygous for high (HH) and low (LL) expression of PRPF31 and a symptomatic and asymptomatic individual. This information will also be useful for formulating and testing potential future gene or gene-based therapies.

No description available.

No description available.

No description available.

No description available.

No description available.

No description available.