Search

The aim of this thesis is to explore how modern computing technology can advance clinical ophthalmology. In particular it will investigate the potential for computer programming in developing measures for assessment of ophthalmic disease. Such measures could be used to advance clinical care or as research tools, providing objective outcomes to power more robust clinical trials. Key aspects of clinical ophthalmic assessment are measurement of visual function and observation of anatomical structures. To provide a full demonstration of the scope for computing this thesis will demonstrate innovative measures for both peripheral and central vision as well as image analysis algorithms applicable to structures from the anterior of the eye to the posterior. For peripheral vision testing we investigated a novel technique for assessment of young children using computer games to hold their attention. Our device demonstrated the potential for drastically improving compliance to obtain useful and accurate visual field results. For central vision assessment we developed a prototype visual function tester for patients with macular degeneration. Specifically we were able to demonstrate the feasibility of programming an iPad© such that patients with macular degeneration could perform and physically respond satisfactorily when self testing their vision. Combining advances in electronics and computing we present data to support the potential for using iris recognition cameras to image eyes of children with diseases that cause their corneas to become cloudy and lose transparency. We were subsequently able to apply our own algorithms to produce a score for the opacification of the cornea that would be useful in quantifying these diseases. Finally, with regard to image analysis applications, we describe the development and validation of algorithms for assessment of posterior capsule opacification (re-growth of cataract) and for haemorrhage assessment. All programming and scientific development was done by the author and this thesis therefore presents the potential for computer software development by an ophthalmologist as a model for the invention of new and novel measures at the interface of computing and ophthalmology.

Communication protocols play a central role in modern telecommunication systems. Protocols are developed to assure that a given service by a system can be provided to an end user in a reliable and efficient manner. The ever-growing influence of telecommunication and distributed systems over daily life require the protocols developed to be free from design errors. One can easily imagine the consequences of protocol failure in the automatic teller machines (ATM), the failure of protocols while making phone calls or failure of protocols in safety critical system such as nuclear power plants and aeroplanes. Any subtle design errors in these protocols can lead to expensive and dangerous consequences. The traditional methodologies such as simulation and design by intuition of protocols are already showing weaknesses and can no longer be relied upon in order to develop modern large complex communication and distribution systems. The formal methods are being deployed in the industry and research institute to develop protocols that are highly reliable and ideally do not contain any errors. Generally, speaking these formal methods can be divided into two categories namely analytic versus synthetic. The analytical methods to design protocols are relatively simple but by no mean guarantee the generation of protocols that are free from design errors. The methods used can only detect errors after the protocol has been designed by means of simulation and validation using automated tools. The synthetic methods allow the generation of protocols that are free from design errors using advanced discrete mathematical concepts and state theory. The complexity of the methods proposed makes it extremely difficult to use them for the protocol of even simplistic natures. The methods proposed so far as shown their usability to the protocol of academic nature only. A semi formal graphical method proposed by this study to design protocol suggest a synthesis method that uses formal graphical language to synthesis the protocols. It overcomes a number of difficulties that are present in formal synthetical methods and early results show that the method can be applied to the protocol of realistic nature. This method also has some weaknesses but it is hoped that future research in this area will overcome some of these.

The mucopolysaccharidoses (MPSs) are a group of rare heterogeneous inherited lysosomal storage disorders. Eye features include corneal clouding, retinopathy, and optic neuropathies, which can be sight threatening and impact quality of life. Over the past few decades systemic treatments have improved systemic outcomes and life span. Therefore, understanding the ocular phenotype and effects of systemic treatments on MPS eye features, is now of greater importance. The majority of studies in this area are small, retrospective or utilise subjective measures of the ocular features in MPS, making findings difficult to interpret. This prospective observational study aims to better understand the ocular phenotype in MPS and provide insights into the effects of systemic treatments. Participants were recruited from paediatric ophthalmology clinics at Manchester Royal Eye Hospital. Those over 3-years of age with any type of MPS were eligible. Follow-up was planned for 4-12 month intervals and at each review where possible participants underwent clinical assessment (including visual acuity, intraocular pressure measurement, and slit lamp examination) and ocular imaging (iris camera, Pentacam, Optos wide-field fundus imaging and optical coherence tomography (OCT)). Participants with MPSI, II, IV and VI were recruited. Corneal clouding was observed in MPSI, IV and VI, and retinopathy and optic nerve abnormality in MPSI, II and VI. Visual acuity did not correlate with grade of corneal clouding. Over a median 22 month follow-up, on average stability of corneal clouding was detected using clinical grading and when measured with the iris camera (COM) score, in all types of MPS studied. Pentacam average change in densitometry scores were higher in MPSI (7.7) and MPSVI (7.2), which may suggest overall progression of corneal clouding in these groups. Retinal and optic nerve changes also appear to be stable clinically and on widefield fundus imaging and OCT in the majority of participants with MPSI, II, IV and VI. Variability in the ocular phenotype has been observed. This variation may be due to genetic and environmental factors. This may also in part be due to small numbers of participants with certain types of MPS in this cohort. There is also variation in findings with different measures of the ocular phenotype utilised, which may be due to subjective nature of some assessments and the different areas of tissues assessed with different imaging modalities. Continuing to follow up the participants in this cohort and further large prospective studies with longer follow-up, utilising objective measures of the ocular phenotype, and studying a range of types of MPS, will provide useful insights into the progression of the eye features of MPS over time and the effects of systemic treatments.

The cornea is the thin, transparent, and avascular connective tissue that lines the front of the eye. As the most anterior surface and a major refractive component of the eye, it allows light transmission to the rest of the eye. Corneal transparency is a general indicator of corneal health. Diseases and some corneal treatments can alter the clarity of the cornea. Assessment of corneal haze in healthy and diseased eyes will aid in the understanding of how diseases progress and the impact of treatment outcomes. Such information will be informative and useful in a clinical setting to assist disease prognosis for the best possible outcome for patients. The iris camera is a specialised device that employs near-infrared light to capture multiple images in seconds which can be analysed to generate a quantitative (objective) estimate of corneal haze. The objective of this project was to investigate corneal haze and its impact on corneal clarity by evaluating changes that take place in health and disease. Specialised algorithms were used to analyse images from an iris camera. Firstly, we assessed changes in corneal transparency over time in a corneal disease with no treatment and compared outcomes with healthy individuals. Corneal diseases and treatments associated with haze were identified, and 163 participants were enrolled and followed up through treatment. We found that corneal clarity changes between patients with the corneal disease (keratoconus) and healthy participants were significantly different. When we compared the outcomes between the two groups, we found that corneal clarity in the healthy participants remained the same with no change while clarity in patients with keratoconus deteriorated. When we compared the outcome of corneal cross-linking on corneal clarity in patients with keratoconus over a certain period, we discovered that the haze level changed (declined) over time, suggesting an improvement. Likewise, in advanced keratoconus, penetrating keratoplasty (PK) improves vision as the last resort when all other remedies prove ineffective against disease progression. The corneal clarity outcomes post-PK again showed improvement, with an observed decline in the haze over time. The corneal response to several treatment methods and disease progression are dependent on a host of factors and will affect corneal clarity differently. With its image analysis software, the iris camera provides an objective means of monitoring corneal clarity in health, disease and assessing treatment modalities.

This thesis presents the outcomes of research work whose focus lies on the development and investigation of new approaches in 3 technologies: perimetry, ocular coherence tomography and retinal implants for the purposes of vision restoration. The thesis is separated in 3 sections: Section 1 consists of a series of studies related to perimetry and glaucoma. The initial investigation of the rate of VF deterioration in patients with different stages of glaucomatous loss showed the relatively low proportion of patients with rapid progression highlighting the effectiveness of current treatment plans. However, the large proportion of patients with advanced field loss presented for the first time emphasised the need for earlier detection of the disease. The following studies focused on the development and evaluation of a new computer based visual field (VF) self administered test for enhanced case-finding of eyes with glaucomatous VF defects. Online VF self-tests were identified and undergone usability evaluation to identify design and testing features that are more attractive to users; such a feature, for example, was the presentation of multiple stimuli. The results of that study were implemented into the design of the new test: a multiple stimulus supra-threshold (i.e. 10dB above age-matched normal threshold) algorithm to test a 20 location subset of the 24-2 pattern with a multisampling (i.e. 3 seen or missed, maximum 5 trials) technique. The performance evaluation of the new test reported specificity at 97% and sensitivity at 85-90%, depending on the stage of loss. Section 2 introduces the new technology of non invasive angiography by means of ocular coherence tomography. Five studies utilising ocular coherence tomography angiography (OCTA) report and discuss the newly acquired knowledge of the eye's vasculature in pathologies, such as diabetic retinopathy or age-related macular degeneration. OCTA proved to be a quick and non invasive mean for angiographic analysis, although drawbacks, such as artefacts or limitations in image acquisition and processing, make it harder to introduce OCTA as a replacement of fundus fluorescein angiography (i.e. the clinical standard for the visualisation of the eye's vasculature). At the end of the section, the potentials of this new technology and future research pathways are thoroughly discussed. The last section describes a clinical trial that evaluates the safety and efficacy of the Argus II retinal implant in patients with advanced non-exudative age related macular degeneration. The presented results emphasise the structural alterations that the implantation caused and the inability of the functional testing to detect any benefits of the Argus II system to the implanted eyes. The section concludes with a critical review of the study's protocol, highlighting its strengths and weaknesses.

Glaucoma can be defined as group of optic neuropathies associated with structural changes to the optic nerve head and the presence of irreversible visual field loss that may derive from various pathological mechanisms. One of the commonest types of glaucoma is primary open angle glaucoma (POAG). Evaluation of the visual field is an essential component in the diagnosis and management of glaucoma. Visual field loss in POAG is often asymmetric both within and between eyes. This thesis extracts samples from a large database of visual field records collected at Manchester Royal Eye Hospital to measure the discriminatory power of various asymmetry indices in the detection of POAG.The first investigation describes and evaluates 3 new Sup/Inf hemifield asymmetry indices; hemifield mean difference, hemifield standard deviation and the number of asymmetric test pairs, for detecting POAG and compares the performance of these new indices with the well-established Glaucoma Hemifield Test (GHT). A good performance was observed for the new indices. GHT can fail to detect significant asymmetry, detected by hemifield standard deviation and number of asymmetric test pairs, when an early defect crosses the GHT sector boundaries. The second investigation looked at the performance of R/L asymmetry indices and whether these indices are superior to the previously investigated Sup/Inf hemifield indices. This study found that, the new R/L asymmetry indices did not perform better than worse eye uniocular hemifield asymmetry indices in differentiating between normal and POAG eyes. Their performance was similar to that of combined uniocular indices. The third investigation describes and measures the discriminatory power of 2 indices that quantify the clustering of Sup/Inf hemifield asymmetric test locations. The final experimental chapter determines the discriminatory power of a multi- factorial model that combines a range of asymmetry indices. In a sample selected to critically evaluate early functional loss the model correctly classified 89% of controls and 80% of POAG cases and performed better than GHT and single asymmetry analyses.