CHAPTER ONE

INTRODUCTION

1. 1. Background of the Study

Retinal diseases are on the rise due to the increase of diabetic population and increased longevity. A good example of the effect of age on retinal health is age-related macular degeneration (AMD): in 2020, the number of people with this disease is projected to be 196 million and is expected to grow to 288 million by 2040 (Wong et al., 2014). Glaucoma is a progressive optic neuropathy, which is the leading cause of blindness in industrialized countries and ocular hypertension is the main risk factor for glaucoma (2016).

There are a number of retinal diseases reported so far such as Arteriosclerotic retinopathy (AR), Central retinal vein occlusion (CRVO), Central retinal artery occlusion (CRAO), Branch retinal vein occlusion (BRVO), Branch retinal artery occlusion (BRAO), Coat’s disease (CD), Hemi-Central Retinal Vein Occlusion (HRVO), Histoplasmosis (HP), Hypertensive retinopathy (HR), Choroidal neovascularization (CNV) and diabetic retinopathy (DR), which may even lead to permanent vision loss. Out of these, age related macular degeneration (AMD) and diabetic retinopathy have been identified as the most significant (Pardue et al., 2018).  

The restrictions of the human eye and inadequacy of the conventional techniques to diagnose the various types of retinal diseases accurately and early in advance are the major challenges faced by the present-day ophthalmologists in the correct treatment of the patients (Valverde et al., 2012). At present, highly sophisticated and dependable diagnostic imaging techniques such as ‘Fluorescent Retinal Angiography (FRA) and Optical Coherence Tomography (OCT) etc. are very popular. A countless number of machine learning approaches such as the artificial neural network (ANN), K-nearest neighbor algorithm, support vector machine (SVM) and Naive Bayes classifier (NBC) are incorporated to improve the prediction accuracy towards the detection of retinal diseases from the images (Ravudu et al., 2012).

Traditional classification approaches depend on feature extraction and feature classification techniques designed for the specific problem based on the available knowledge of the field. Most of the algorithms used in this area encounter the challenge of having the only insufficient number of datasets for training the model through conventional machine learning techniques (Li et al., 2019).

The launching of ‘deep learning CNN’ based algorithms makes evolutionary changes in the approach by directly identifying features from the training data without the categorical elaboration on feature extraction and classification. However, deep learning-based models are found to improve their efficiency on vigorous training using a large number of datasets (Karthikeyan et al., 2017). This study aims to use deep learning algorithm, CNN to classify various of retinal diseases using images.

1.2     Statement of the Problem

Retinal diseases can cause loss of vision in the human eye. Retinal disease can however be diagnosed by an ophthalmologist and be treated, the main goals of treatment are to stop or slow disease progression and preserve, improve or restore one’s vision. In many cases, damage that has already occurred can't be reversed, making early detection very important. The early detection of retinal disease has been a major challenge in the area of medical sciences; thus, a retinal disease classification system is been proposed here to aid in the early and easy detection of retinal problems using deep learning to train a classification model from a dataset of fungus retinal disease images. The proposed system would make prediction of retinal disease using a mobile device camera to capture input images.

1.3     Aim and Objectives of the Study

The aim of this research is to develop a retinal disease classification system, that will use a deep learning (DL) algorithm known as Convolutional Neural Network (CNN) to classify retinal diseases through images.

The objectives of the study are to;

1. i. Design a retinal disease classification system using Unified Model Language (UML) tools.
2. Develop a deep learning model for retinal disease classification.
3. Reduce the time frame it takes medical professionals in the detection of retinal diseases by the deployment of the of the trained model.
4. Evaluate the developed system for performance.

1.4     Significance of the Study

The Deep Learning Based retinal diseases classification will be of significant benefits to the medical sector, via the deployment of the developed software along medical professionals in charge of the diagnosis of retinal diseases, which will turn bring about more ease and less time consumption in the diagnosis of retinal defects.

The software is relevant in the following ways:

1. The designed software will aid in carrying eye(retinal) disease diagnosis.
2. The designed software will reduce the time frame it would take an ophthalmologist to diagnose a retinal disease.
3. The system will serve as a conduit for further research by scholars and academicians.
4. The designed software will also be relevant to any individual or agency which will like to carry out a survey on performance of CNN in retinal disease classification.

1.5     Scope of the Study

This research work covers the development of an improved retinal disease classification using convolutional neural network to train a model that classifies retinal diseases via images.

1.6 Definition of Terms

Artificial Intelligence (AI): This is intelligence demonstrated by machines, unlike the natural intelligence displayed by humans and animals, which involves consciousness and emotionality.

Convolutional Neural Network (CNN): This is a Deep Learning algorithm which can take in an input image, assign importance (learnable weights and biases) to various aspects/objects in the image and be able to differentiate one from the other.

Data analysis: This is a process of inspecting, cleansing, transforming and modeling data with the goal of discovering useful information, informing conclusion and supporting decision-making.

Data Argumentation:   Is use in deep learning to regularize, normalize the information on data to be trained.

Data Notation:    This is the process of labeling data to make it usable for machine learning.

Dataset: A data set is a collection of data. In the case of tabular data, a data set corresponds to one or more database tables, where every column of a table represents a particular variable, and each row corresponds to a given record of the data set in question.

Deep Learning (DL): This is a subfield of machine learning concerned with algorithms inspired by the structure and function of the brain called artificial neural networks.

Machine Learning (ML): This is the study of computer algorithms that improve automatically through experience and by the use of data.

Mobile App:  A mobile app or mobile application is a computer program or software application designed to run on a mobile device such as a phone-tablet or watch.

Model: Is a description or analogy used to help visualize something that cannot be directly observed.

Disease: A disease is an abnormal condition that negatively affects the structure or function of all or part of an organism, and that is not due to any immediate external injury. Diseases are often known to be medical conditions that are associated with specific signs and symptoms.

Multilayer Perceptron: This refers to fully connected networks, that is, each neuron in one layer is connected to all neurons in the next layer.

Neural Networks (NN): This is a series of algorithms that endeavors to recognize underlying relationships in a set of data through a process that mimics the way the human brain operates.

Noisy Data: Noisy data is data that is corrupted, or distorted, or has a low Signal-to-Noise Ratio. Improper procedures to subtract out the noise in data can lead to a false sense of accuracy or false conclusions.

Perceptron: This refers to a linear classifier (binary) use in supervised learning. It helps to classify the given input data.

Retina: The retina is a thin layer of tissue that lines the back of the eye on the inside. It is located near the optic nerve. The purpose of the retina is to receive light that the lens has focused, convert the light into neural signals, and send these signals on to the brain for visual recognition.

Software: These are set of instructions that are designed to solve a specific problem. Software is made up of data and instructions. There are two categories of software; System software and Application software.

System: A system is a set of detailed methods, procedures and routines created to carry out a specific activity, perform a duty, or solve a problem

Tensor-Flow: is a free and open-source software library for dataflow and differentiable programming across a range of tasks. It is a symbolic math library, and is also used for machine learning applications such as neural networks.

Unified Modeling Language: The Unified Modeling Language is a general-purpose, developmental, modeling language in the field of software engineering that is intended to provide a standard way to visualize the design of a system.