Abstract:The analysis of retinal blood vessels present in fundus images, and the addressing of problems such as blood clot location, is important to undertake accurate and appropriate treatment of the vessels. Such tasks are hampered by the challenge of accurately tracing back problems along vessels to their source. This is due to the unresolved issue of distinguishing automatically between vessel bifurcations and vessel crossings in colour fundus photographs. In this paper, we present a new technique for addressing this problem using a convolutional neural network approach to firstly locate vessel bifurcations and crossings and then to classifying them as either bifurcations or crossings. Our method achieves high accuracies for junction detection and classification on the DRIVE dataset and we show further validation on an unseen dataset from which no data has been used for training. Combined with work in automated segmentation, this method has the potential to facilitate: reconstruction of vessel topography, classification of veins and arteries and automated localisation of blood clots and other disease symptoms leading to improved management of eye disease.

Opisthorchis viverrini infection was found to be highly prevalent in 3 riverside villages (Ang Svay Chek A, B, and C) of the Prey Kabas District, Takeo Province. This area is located in the southern part of Cambodia, where the recovery of adult O. viverrini worms was recently reported. From May 2006 until May 2010, fecal examinations were performed on a total of 1,799 villagers using the Kato-Katz thick smear technique. In the 3 villages, the overall positive rate for helminth eggs ranged from 51.7 to 59.0% (av. 57.4%), and the percentage positive for O. viverrini was 46.4-50.6% (47.5%). Other helminths detected included hookworms (13.2%), echinostomes (2.9%), Trichuris trichiura (1.3%), Ascaris lumbricoides (0.6%), and Taenia spp. (0.06%). The prevalence of O. viverrini eggs appeared to reflect a lower infection in younger individuals (<20 years) than in the adult population (>20 years). Men (50.4%) revealed a significantly higher (P=0.02) prevalence than women (44.3%). The Ang Svay Chek villages of the Prey Kabas District, Takeo Province, Cambodia have been confirmed to be a highly endemic area for human O. viverrini infection.

There has been a significant increase in incidence of cases of Acanthamoeba keratitis presenting to our institution. The type of contact lens solution and the use of silicon hydrogel lenses combined with extended overnight wear may play a role; however, the significance is unclear given the small numbers for analysis. Further study of incidence and patient characteristics is warranted to identify risk factors and causes for the rising incidence.

BackgroundCentral vision, critical for everyday tasks such as reading and driving, is impacted by age-related changes in the eye and by diseases such as age-related macular degeneration. The detection of changes in macular function is therefore important. The Radial Shape Discrimination (RSD) test measures the threshold at which distortions in a radial frequency pattern can be detected and there is evidence that it is more sensitive to macular pathology than visual acuity (VA). It also provides a more quantitative measure of macular function than the commonly available Amsler grid. Recently, handheld versions of the test (hRSD) in which stimuli are presented on mobile devices (e.g., Apple iPod Touch, iPhone) have been developed. We investigated the characteristics of the hRSD test in healthy participants.MethodsData were collected using both three-alternative forced choice (3AFC) and 4AFC versions of the hRSD test, presented on an Apple iPod Touch. For the 3AFC version, data from a single test session were available for 186 (72 male; mean ± SD age 42 ± 17y; range 16–90y) healthy participants. Test-retest data were available for subgroups of participants (intra-session: N = 74; tests approximately 2 months apart: N = 30; tests 39 months apart: N = 15). The 3AFC and 4AFC versions were directly compared in 106 participants who also completed a usability questionnaire. Distance and near VA and Pelli Robson Contrast Sensitivity (CS) data were collected and undilated fundoscopy performed on the majority of participants.ResultsMean (±SD) 3AFC hRSD threshold was −0.77 ± 0.14 logMAR, and was statistically significantly correlated with age (Pearson r = 0.35; p < 0.001). The linear regression of hRSD threshold on age had a slope of +0.0026 compared to +0.0051 for near VA (which also correlated with age: r = 0.51; p < 0.001). There were no statistically significant differences in hRSD thresholds for any of the test-retest subgroups. We also observed no statistically significant difference between 3AFC (−0.82 ± 0.11 logMAR) and 4AFC (−0.80 ± 0.12 logMAR) hRSD thresholds (t = 1.85, p = 0.067) and participants reported excellent test usability with no strong preference expressed between the 3AFC and 4AFC versions of the test.DiscussionThe 3AFC hRSD thresholds we report are consistent with a number of previous studies, as is its greater stability in ageing compared to VA. We have also shown that in the absence of pathology, thresholds are stable over short and long timescales. The 4AFC thresholds we have reported provide a baseline for future investigations, and we have confirmed that 3AFC and 4AFC thresholds are similar, providing a basis of comparisons between studies using the different versions. As the hRSD test is easy to use and relatively inexpensive, clinical studies are now required to establish its ability to detect and monitor macular pathologies.

No abstract