A Study of Cloud Computing for Retinal Image Processing through MATLAB

A Study of Cloud Computing for Retinal Image Processing through MATLAB

S. K. Maharana, Ganesh Prabhakar P., Amit Bhati
Copyright: © 2013 |Pages: 11
DOI: 10.4018/978-1-4666-3994-2.ch006
OnDemand:
(Individual Chapters)
Available
$37.50
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

With the ability of MATLAB to run in the cloud environment the authors analyse a retinal image efficiently. A quick and on-the-fly image processing is the cause for leveraging the ability of Cloud with the classic computing power of MATLAB. This cloud-based image processing has significantly saved on the cost of procuring resources and it has processed an image in a few seconds. This revolutionary change in computing power has not only eased the life of engineering community but has demonstrated an inertia to help the common man through its language of computing. The objective of this particular piece of work is to leverage the ubiquitous cloud features to process the images taken from retina portion through a hi-fidelity algorithm. The reflection of the computing ability of complex mathematical equations, SaaS (Soft as a Service) architecture of cloud and tools developed in Windows Azure platform has come out as research findings such as the spatial modelling of a diseased portion of a retina. The portion diagnosed with a difficulty undergoes a series of steps based on the algorithm developed by authors and ultimately the original image is transformed into a form with information on the status of the disease in the retina.
Chapter Preview
Top

Introduction

With the ability of MATLAB (http://mathworks.com) to run in the Cloud environment we have attempted to do a complex computation of analyzing a retinal image pretty efficiently. This cloud-based image processing has significantly saved a good amount of cost of procuring expensive resources and it has processed an image in a few seconds. The speed and the accuracy involved in executing this task has come from the cloud based set-ups and a few tools we have developed. The objective of this paper is to demonstrate how ubiquitous cloud services and features have been leveraged to process the images taken from retina portion through a hi-fidelity algorithm. The interest to carry out this research has been derived from medical image processing (Bankman, 2000). It is the confluence of two sciences that behave in completely different ways. While medicine is a science where experience plays a majors role and where the practical use is evident, image processing—as a derivative of applied mathematics—is a more theoretical discipline. Hence, the conditions of this confluence need to be analyzed sophisticatedly. It is known that diabetic retinopathy (American Academy of Opthalmology, n.d.) is a severe and widespread eye disease. In fact, it is the leading cause of legal blindness for the working age population (between 20 and 64) in many places in the globe. Diabetic retinopathy is a complication of diabetes mellitus and its prevalence increases with the duration of the disease. After 15 years of diabetes, the prevalence lies near 98%, so nearly all diabetic patients are affected by this disease after some time. Although not all the forms of the disease coincide with vision alteration, about 2% of the diabetic patients are blind and 10% suffer from vision loss after 15 years of diabetes.

The above reasons have generated lot of interest to develop a fast and accurate method based in cloud so that the processing could be done from anywhere if we have the image is available. The algorithm, the flow of computation and processing of a particular image is carried out using MATLAB through a cloud based user interface.

Complete Chapter List

Search this Book:
Reset