Convolutional Neural Networks for Diabetic Retinopathy

Computer-aided diagnosis of diabetic retinopathy: A review

Glaucoma is a disease of the optic nerve caused by the increase in the intraocular pressure of the eye. Glaucoma mainly affects the optic disc by increasing the cup size. It can lead to the blindness if it is not detected and treated in proper time. The detection of glaucoma through Optical Coherence Tomography (OCT) and Heidelberg Retinal Tomography (HRT) is very expensive. This paper presents a novel method for glaucoma detection using digital fundus images. Digital image processing techniques, such as preprocessing, morphological operations and thresholding, are widely used for the automatic detection of optic disc, blood vessels and computation of the features. We have extracted features such as cup to disc (c/d) ratio, ratio of the distance between optic disc center and optic nerve head to diameter of the optic disc, and the ratio of blood vessels area in inferior-superior side to area of blood vessel in the nasal-temporal side. These features are validated by classifying the normal and glaucoma images using neural network classifier. The results presented in this paper indicate that the features are clinically significant in the detection of glaucoma. Our system is able to classify the glaucoma automatically with a sensitivity and specificity of 100% and 80% respectively.

Automated Diagnosis of Glaucoma Using Digital Fundus Images

Purpose To model the photochemical kinetics of corneal cross-linking with riboflavin (Rf) and confirm the model through measured oxygen concentration experiments under varying energy input conditions by UV-A irradiance and temperature modulation in ex vivo porcine cornea. Methods A theoretical model was developed to describe the corneal cross-linking photochemical kinetics of Rf. After instillation with drops of Rf solution in distilled water, de-epithelialized porcine corneas were exposed to 365-nm ultraviolet light (UV-A) under varying irradiance and temperature. Oxygen concentration in the cornea at a known depth was monitored during UV-A illumination with a dissolved oxygen fiberoptic microsensor. Data from the oxygen experiments were used to confirm the model. Results On the basis of the known chemical reactions and diffusion rates of Rf and oxygen into the cornea, the authors developed a theoretical model consistent with corneal oxygen consumption experimental results during UV-A irradiation under different conditions. Oxygen concentration in the cornea is modulated by UV-A irradiance and temperature and quickly decreased at the beginning of UV-A exposure. The time-dependence of both Type-I and Type-II photochemical mechanisms in corneal cross-linking with Rf are discussed. Conclusions Using a chemical kinetics modeling approach, the authors developed a simple model that is in agreement with their experimental results on oxygen consumption in the cornea during corneal cross-linking with Rf. It is suggested that the main photochemical kinetics mechanism is the direct interaction between Rf triplets and reactive groups of corneal proteins, which leads to the cross-linking of the proteins mainly through radical reactions.

Photochemical kinetics of corneal cross-linking with riboflavin.

Diabetes is a chronic end organ disease that occurs when the pancreas does not secrete enough insulin or the body is unable to process it properly. Over time, diabetes affects the circulatory system, including that of the retina. Diabetic retinopathy is a medical condition where the retina is damaged because fluid leaks from blood vessels into the retina. Ophthalmologists recognize diabetic retinopathy based on features, such as blood vessel area, exudes, hemorrhages, microaneurysms and texture. In this paper we review algorithms used for the extraction of these features from digital fundus images. Furthermore, we discuss systems that use these features to classify individual fundus images. The classifications efficiency of different DR systems is discussed. Most of the reported systems are highly optimized with respect to the analyzed fundus images, therefore a generalization of individual results is difficult. However, this review shows that the classification results improved has improved recently, and it is getting closer to the classification capabilities of human ophthalmologists.

Algorithms for the Automated Detection of Diabetic Retinopathy Using Digital Fundus Images: A Review

Diabetic retinopathy (DR) is caused by damage to the small blood vessels of the retina in the posterior part of the eye of the diabetic patient. The main stages of diabetic retinopathy are non-proliferate diabetes retinopathy (NPDR) and proliferate diabetes retinopathy (PDR). The retinal fundus photographs are widely used in the diagnosis and treatment of various eye diseases in clinics. It is also one of the main resources for mass screening of diabetic retinopathy. In this work, we have proposed a computer-based approach for the detection of diabetic retinopathy stage using fundus images. Image preprocessing, morphological processing techniques and texture analysis methods are applied on the fundus images to detect the features such as area of hard exudates, area of the blood vessels and the contrast. Our protocol uses total of 140 subjects consisting of two stages of DR and normal. Our extracted features are statistically significant (p?<?0.0001) with distinct mean?±?SD as shown in Table 1. These features are then used as an input to the artificial neural network (ANN) for an automatic classification. The detection results are validated by comparing it with expert ophthalmologists. We demonstrated a classification accuracy of 93%, sensitivity of 90% and specificity of 100%.

Automated Identification of Diabetic Retinopathy Stages Using Digital Fundus Images

The human body temperature is a good health indicator. All objects emit thermal radiation as a function temperature and wavelength for all wavelengths. The wavelength of infrared rays lies between visible and microwave radiations ranging between 700 nm to 0.1 mm. Infrared (IR) imaging is relatively inexpensive, noninvasive and harmless. Nowadays, it is widely used in the medical field for diagnosis. In this work, we have applied image processing techniques on the IR images of the eye for the analysis of the ocular surface temperature (OST) of the normal subjects of three categories (young, middle and old ages). In our study, 67 IR normal images were analyzed. Two parameters, average ocular temperature and the temperature deviation were proposed to study the variability of OST in different normal category subjects. Our study shows that, the two parameters proposed, show distinct ranges for different groups with `p' values less than 0.05.

Manjunath Kagathi

Papers

Automated Identification of Diabetic Retinopathy Stages Using Digital Fundus Images

Analysis of Normal Human Eye with Different Age Groups using Infrared Images