Optic-Net: A Novel Convolutional Neural Network for Diagnosis of Retinal Diseases from Optical Tomography Images

Abstract

Diagnosing different retinal diseases from Spectral Domain Optical Coherence Tomography (SD-OCT) images is a challenging task. Different automated approaches such as image processing, machine learning and deep learning algorithms have been used for early detection and diagnosis of retinal diseases. Unfortunately, these are prone to error and computational inefficiency, which requires further intervention from human experts. In this paper, we propose a novel convolution neural network architecture to successfully distinguish between different degeneration of retinal layers and their underlying causes. The proposed novel architecture outperforms other classification models while addressing the issue of gradient explosion. Our approach reaches near perfect accuracy of 99.8% and 100% for two separately available Retinal SD-OCT data-set respectively. Additionally, our architecture predicts retinal diseases in real time while outperforming human diagnosticians.

A novel deep learning conditional generative adversarial network for producing angiography images from retinal fundus photographs

“…The generator block consists of two generator modules, the fine and coarse generators, which are designed in a U-shaped encoder-decoder manner. The coarse generator is comprised of a reflection+padding block, three convolution (Conv)+batch normalization (BN)+leaky rectified linear units (ReLU), and four novel residual blocks 44,45 (ResBlk), followed by two transpose convolution (Deconv), one reflection+padding, one Conv, and an output activation layers ( Fig. 1B-left), and is responsible for generating coarse and global structures of the FA image such as the structures of the macula, optic disc, color, contrast, and brightness.…”

“…Deep learning conditional generative adversarial network. This study proposes a new conditional generative adversarial network (GAN) comprising of a novel residual block 44,45 for producing realistic FA from retinal fundus images. We use two generators ( G fine and G coarse ) in the proposed network, as illustrated in Fig.…”

“…In order to address the inefficiency in utilizing generic deep learning models in ophthalmology, our team has developed effective architectures [44][45][46] for retinal disease diagnosis. Our proposed architectures are capable of mapping ophthalmic images (e.g.…”

“…Given the ability to use our previous deep learning architectures [44][45][46] to produce feature rich latent manifolds from input ocular structural images, we sought to explore the ability to map paired fundus photographs and FA images onto a shared latent manifold in which the retina vasculature from both domains share similar feature representations. This approach has its roots in the recently introduced Generative Adversarial Networks (GANs) 27,[50][51][52] in the field of deep learning.…”

RV-GAN: Segmenting Retinal Vascular Structure in Fundus Photographs using a Novel Multi-scale Generative Adversarial Network

“…As a result, it extracts and preserves depth and spatial features while forward propagating the network. Recent advancement in retinal image classification has shown that combining separable convolutional layers with dilation allows for more robust feature extraction [19]. We design two unique residual identity blocks, for our generators and discriminators, as illustrated in Fig.…”