In this dissertation, I investigated computing algorithms for automated retinal blood vessel detection. Changes in blood vessel structures are important indicators of many diseases such as diabetes, hypertension, etc. Blood vessel is also very useful in tracking of disease progression, and for biometric authentication. In this dissertation, I proposed two algorithms to detect blood vessel maps in retina. The first algorithm is based on integration of a Gaussian tracing scheme and a Gabor-variance filter. This algorithm traces the large blood vessel in retinal images enhanced with adaptive histogram equalization. Small vessels are traced on further enhanced images by a Gabor-variance filter. The second algorithm is called a radial contrast transform (RCT) algorithm, which converts the intensity information in spatial domain to a high dimensional radial contrast domain. Different feature descriptors are designed to improve the speed, sensitivity, and expandability of the vessel detection system. Performances comparison of the two algorithms with those in the literature shows favorable and robust results. Furthermore, a new performance measure based on central line of blood vessels is proposed as an alternative to more reliable assessment of detection schemes for small vessels, because the significant variations at the edges of small vessels need not be considered. The proposed algorithms were successfully tested in the field for early diabetic retinopathy (DR) screening. A highly modular code library to take advantage of the parallel processing power of multi-core computer architecture was tested in a clinical trial. Performance results showed that our scheme can achieve similar or even better performance than human expert readers for detection of micro-aneurysms on difficult images.

Magister Public Health - MPH / BACKGROUND AND RATIONALE: The Cape Town Metro District Health Service (MDHS) has introduced a Diabetic RetinopathyScreening (DRS) programme incorporating retinal fundal photography in diabetic services at primary health care (PHC) facilities. Hitherto, coverage of the DRS programme has been less than optimal in part due to volumes of diabetic patients attending PHC facilities. The aim of this study was to identify possible sub-groups of patients, attending the Cape Town DRS Programme, who are at most risk of diabetic retinopathy and might be prioritised for early diabetic retinopathy detection and subsequent sight-saving treatment. METHODOLOGY: A case-control study of risk factors for treatment-requiring diabetic retinopathy was conducted. This research sampled participants from the DRS programme provided by the MDHS eye care team to Type II diabetics attending public PHC facilities within the Klipfontein and Mitchells Plain Sub-Districts. Based on fundal images, cases were selected as those requiring ophthalmological treatment; and controls (three matched per case by area of residence) as those judged as not requiring ophthalmological treatment for diabetic retinopathy. Data on possible risk factors (clinical, laboratory) were extracted from the patients' folders. RESULT: The study included 453 participants, of whom 113 (24.9%) were cases and 340 (75.1%) were controls. Three factors were significantly associated with treatment-requiring diabetic retinopathy on multivariate analysis: Insulin dependency (OR of 2.96, 95% CI: 1.75 – 5.00); duration of diabetes of more than 10 years (OR of 3.44, 95% CI: 2.06 – 5.74) and sustained hyperglycaemia over the past six months (OR of 3.73, 95% CI: 1.69 – 8.22). A screening algorithm combining these criteria had a sensitivity of 61.2% (95% CI: 51.9 – 70.5). CONCLUSION: The findings indicate that a sub-set of patients attending the DRS programme in the Klipfontein and Mitchells Plain Sub-Districts have a greater likelihood of presenting with treatment-requiring diabetic retinopathy. Further research is required to develop a tool that is sufficiently sensitive to safely prioritise patients for fundal screening. / National Research Foundation (NRF)