Comparing the Clinical Viability of Automated Fundus Image Segmentation Methods.

Recent methods for automatic blood vessel segmentation from fundus images have been commonly implemented as convolutional neural networks. While these networks report high values for objective metrics, the clinical viability of recovered segmentation masks remains unexplored. In this paper, we perform a pilot study to assess the clinical viability of automatically generated segmentation masks in the diagnosis of diseases affecting retinal vascularization. Five ophthalmologists with clinical experience were asked to participate in the study. The results demonstrate low classification accuracy, inferring that generated segmentation masks cannot be used as a standalone resource in general clinical practice. The results also hint at possible clinical infeasibility in experimental design. In the follow-up experiment, we evaluate the clinical quality of masks by having ophthalmologists rank generation methods. The ranking is established with high intra-observer consistency, indicating better subjective performance for a subset of tested networks. The study also demonstrates that objective metrics are not correlated with subjective metrics in retinal segmentation tasks for the methods involved, suggesting that objective metrics commonly used in scientific papers to measure the method's performance are not plausible criteria for choosing clinically robust solutions.

Intravitreal bevacizumab injection alone or combined with macular photocoagulation compared to macular photocoagulation as primary treatment of diabetic macular edema.

BACKGROUND/AIM: Within diabetic retinopathy (DR), diabetic macular edema DIE) is one of the leading causes of the loss of visual acuity. The aim of this study was to determine the efficacy of the intravitreal vascular endothelial growth factor (VEGF) inhibitor application alone or combined with macular focal/grid lasephotocoagulation compared with laser treatment alone. METHODS: This prospective randomized clinical trial included 72 patients (120 treated eyes) with varying degrees of DR and DME. The DME treatment included intravitreal VEGF inhibitor bevacizumab (Avasting) application, with and without laser treatment. Bevacizumab (1.25 mg/0.05 mL) was administered intravitreally in 4-6-week intervals. Laser is applied 4-6 weeks after last dose of the drug as a part of combined treatment, or as the primary treatment. RESULTS: The mean reduction in central macular thickness (CMT) for the eyes (n = 31) treated with bevacizumab alone was 162.23 .rm, for the eyes (n = 53) treated with combined treatment the mean reduction in CMT was 124.24 pm, both statistically significant at p < 0.001. Laser macular photocolagulation as a part of combined treatment (in 53 eyes) significantly contributed to the CMT reduction, based on the paired t-test results (366.28 vs. 323.0 pLm at p < 0.05). In our study, the mean visual acuity improvement of 0.161 logMAR was achieved in the group of eyes treated with bevacizumab alone, and 0.093 logMAR in the group with combined treatment, both statistically significant atp < 0.05. The effect of laser photocolagulation alone on visual acuity and'CMT was not statistically significant. CONCLUSION: Treatment with bevacizumab alone or within combined treatment is more effective in treating DME than conventional macular laser treatment alone, both anatomically and functionally.

PHARMACOLOGICAL INTRAVITREAL TREATMENT FOR MACULAR EDEMA IN BRANCH RETINAL VEIN OCCLUSION - THREE-MONTH RESULTS.

INTRODUCTION: Macular edema is the main cause of visual loss in patients with branch retinal vein occlusion. Macular edema is initially reversible, but over time, permanent loss ofvision occurs from structural damage to the macula. For this reason, there is a need for more rapid and effective treatments than laser photocoagulation which has been established as a gold standard. There are several pharmacologic agents which have changed the management of macular edema. MATERIAL AND METHODS: Twenty eyes of 20 consecutive patients of the Department of Eye Diseases, Clinical Center of Vojvodina, in Novi Sad, were enrolled in this prospective, randomized and consecutive study conducted from January 2012 to January 2013. The patients were randomly assigned into two treatment groups, and they were given an intravitreal injection of bevacizumab 1.25 mg/0.05 mL (Avastin®), or triamcinolone acetonid injection 4 mg/0.1 mL (Kenalog®). Reinjections were performed according to the following retreatment criteria a loss of visual acuity or increase in central retinal thickness. RESULTS: Both intravitreal bevacizumab and triamcinolon-acetonid were very effective in reducing macular edema and improving visual acuity in the eyes with macular edema secondary to retinal vein occlusion. The effect of the treatment was more pronounced if it started early after the onset of macular edema. The reported temporary effects of intravitreal triamcinolon-acetonide and bevacizumab could be explained by their clearance from the eye. CONCLUSION: The short-term results of our clinical trial showed that pharmacological intravitreal agents, such as bevacizumab and triameinolon-acetonid, lead to rapid resolution of macular edema and significant improvement of visual acuity.