Deep Feature Vectors Concatenation for Eye Disease Detection Using Fundus Image

Fundus image is an image that captures the back of the eye (retina), which plays an important role in the detection of a disease, including diabetic retinopathy (DR). It is the most common complication in diabetics that remains an important cause of visual impairment, especially in the young and economically active age group. In patients with DR, early diagnosis can effectively help prevent the risk of vision loss. DR screening was performed by an ophthalmologist by analysing the lesions on the fundus image. However, the increasing prevalence of DR is not proportional to the availability of ophthalmologists who can read fundus images. It can lead to delayed prevention and management of DR. Therefore, there is a need for an automated diagnostic system as it can help ophthalmologists increase the efficiency of the diagnostic process. This paper provides a deep learning approach with the concatenate model for fundus image classification with three classes: no DR, non-proliferative diabetic retinopathy (NPDR), and proliferative diabetic retinopathy (PDR). The model architecture used is DenseNet121 and Inception-ResNetV2. The feature extraction results from the two models are combined and classified using the multilayer perceptron (MLP) method. The method that we propose gives an improvement compared to a single model with the results of accuracy, and average precision and recall of 91% and 90% for the F1-score, respectively. This experiment demonstrates that our proposed deep-learning approach is effective for the automatic DR classification using fundus photo data.

Multimodal imaging and adaptive optics in internal limiting membrane dystrophy

An 18-year-old boy presented with a mild blurring of vision in both his eyes for the last 2 years. His best-corrected visual acuity was 20/60 in both the eyes with normal colour vision. Fundus evaluation showed exaggerated shiny reflex from the internal limiting membrane (ILM) with the wrinkled inner retinal surface at posterior pole. No vessel attenuation or retinal pigment epithelium changes were evident. Optical coherence tomography showed thickened ILM, which was appreciated in fundus photo also. The electrodiagnostic tests were normal, and the photoreceptors appeared normal in adaptive optics in the central macula at 2°–4° eccentricity. Among the very few reported cases of Muller cells sheen dystrophy, our patient was the probably the youngest and was picked up by a thoughtful ophthalmologist in detailed clinical examination. Adaptive optics in the case of ILM dystrophy is also not reported earlier.

Diagnosis and Management of Central Serous Chorioretinopathy

Introduction. Abnormalities in the macula is a very significant disorder. One abnormality that can be found is central serous chorioretinopathy (CSC). This case report aims to explain the diagnosis and management of Central Serous Chorioretinopathy. Case presentation. A 32-year-old man, a soldier, came for treatment at hospital on May 2017. Anamnesis of major complaints right eye blurred since, 1 week before hospitalization. From the investigations, in the right eye, the armsler grid examination showed an impression of metamorphopsia. On examination of the fundus photo, an impression of macular area edema with a clear border is obtained. OCT examination revealed that there was elevation of neuroretina tissue in the macula. This patient was managed by laser photocoagulation at the point of leakage of subretinal fluid. Conclusion. The diagnosis of central serous chorioretinopathy is made through history taking, ophthalmological examination, and supporting examinations of armsler grid, photo fundus, OCT and FFA. Laser photocoagulation can be performed, especially in cases of persistent, chronic, recurrent, or occupational needs.

The Guangzhou Twin Eye Study: 2019 Update

The primary aim of the Guangzhou Twin Eye Study (GTES) is to explore the impact that genes and environmental influences have on common eye diseases. Since 2006, approximately 1300 pairs of twins, aged 7–15 years, were enrolled at baseline. Progressive phenotypes, such as cycloplegic refraction, axial length, height and weight, have been collected annually. Nonprogressive phenotypes such as parental refraction, corneal thickness, fundus photo, intraocular pressure and DNA were collected once at baseline. We are collaborating with fellow international twin researchers and psychologists to further explore links with general medical conditions. In this article, we review the history, major findings and future research directions for the GTES.