Improving the structure function relationship in the macula

The macula is the central part of the retina responsible for central vision and can suffer damage from many diseases, including diabetes, macular degeneration and glaucoma. Establishing a relationship between functional measurements, such as perimetry, and structural metrics, such as those obtained through imaging, has proven both clinically appealing and challenging, owing to specific features of this area of the retina. The programme of work presented in this thesis focuses on improving the accuracy of structure-function analyses of the macula as well as the mechanistic understanding of structure-function relationship in both healthy and diseased eyes. The first study revisits and improves previous models quantifying the length of Henle’s fibres. This directly relates to the radial displacement of Retinal Ganglion Cells (RGCs) from their photoreceptors and affects structure-function mapping. The study demonstrated the inaccuracy of previous methods used to displace perimetric stimuli, proposing a correct implementation of these calculations. These results were made available to other researchers in a user-friendly web application. The second study explored how natural positioning of observers in front of imaging and perimetry devices, as well as their fixation and eye movements, affected the precision of macular structure-function mapping. The study analysed data from an eye-tracking perimeter used to test both healthy eyes and patients with glaucoma. An optimal strategy for structure-function mapping was developed and the mapping error introduced by fixation was quantified. The third study used data from an eye-tracking perimeter and the framework of an established neural model of spatial summation to investigate the structure-function relationship in early neural loss in patients with diabetes without diabetic retinopathy, quantified with both imaging and functional tests, including Frequency Doubling Perimetry, standard visual acuity and contrast sensitivity. The fourth study involved the prospective collection of data from healthy observers with perimetric stimuli of different sizes and durations, using custom software. The data were used to develop a computational model of perimetric sensitivity able to reproduce the interaction between spatial and temporal summation in the context of cortical integration and their link to the number of retinal ganglion cells being stimulated. In the fifth study, the methodology and mechanistic framework developed in the previous studies were applied to test the computational model in glaucoma. The model was used to obtain functional estimates of retinal ganglion cell damage from standard automated perimetry data collected in glaucoma patients and healthy age-related controls. The results were correlated with imaging and histology data from previous literature