© IEOM Society
Review of Image Processing Techniques for Detection of Age-
related Macoular Degeneration (ARMD)
Literature Review
Samina Khalid
Computer Science and Information Technology Department, Mirpur University of Science and
Technology
Mirpur AJK Pakistan
noshi_mir@yahoo.com
Dr. M. Usman Ikram
College of E&ME, NUST
Rawalpindi, Pakistan
usman.akram@ceme.nust.edu.pk
Abstract
Detection of retinal drusen is important in the diagnosis of Age-related Macular Degeneration (ARMD).
Automated image processing has the potential to assist in the early detection of ARMD, by changing in blood
vessels and pattern in retina. Age- related macular degeneration cause gradual loss of vision by oxidation of
macula and can cause irreversible vision loss. The main goal of the proposed system is twofold at first it is used to
diagnose the type of age-related macular degeneration either it is dry macular degeneration or wet macular
degeneration, and then further helps to classify the stage of dry macular degeneration into early, intermediate, or
advanced dry ARMD and to automatically detect and segment ARMD without human supervision.
Detection of ARMD is done by using Auto Associative Neural Network (AANN) method and the two
classes of Age-related macular degeneration (dry or wet ARMD), one of which dry macular can be further
classify into three classes, will be classify and diagnose successfully as future work.
Keywords: Fundus Images, OCT Images, Drusen, Macular Degeneration, ARMD.
INTRODUCTION
Age-related macular degeneration (ARMD) the leading cause of worldwide blindness in the elderly age is a bilateral ocular
condition that affects the central area of retina known as the macula. Although the macula comprises only four percent of
retinal area, it is responsible for the majority of useful photonic vision [1]. ARMD is the main cause of the elderly blindness in
developed countries e.g. Australia, United Kingdom, and America. According to a survey approximately 17% of the
participants were diagnosed with ARMD; further, more than 95% of these were aged 60 years and above [2]. In United States
ARMD is also a growing public health problem, almost 11 million, or 7.6% of all Americans are estimated to have ARMD,
and it is the cause of blindness for 54% of all legally blind Americans. ARMD is a major societal problem in terms of
disability and health care costs. For example, severe ARMD reduces the likelihood of employment by 61% and salary by 39%,
while mild ARMD reduces these by 44% and 32% respectively. The estimated annual cost burden from ARMD in the U.S. is
$30 billion (USD) or about 0.3% of gross domestic product [3]. The occurrence of ARMD is expected to double over the next
25 years.
ARMD caused due to deposits of bright lesions called drusen. Drusen are formed at retinal level and could affect eyesight.
Many researches have been done in the field of medical care. The diagnosis of ARMD is typically undertaken through the
inspection of the macula (see Figure:1). Manual recognition and detection of drusen from retinal images is time consuming and
expensive. Moreover, it is subjective and its reproducibility is a concern. To save workload and facilitate large-scale clinical
use, it is important to have a precise, cost effective and efficient system to detect drusen automatically for ARMD diagnosis
[4]. ARMD
Proceedings of the 2015 International Conference on Operations Excellence and Service Engineering
Orlando, Florida, USA, September 10-11, 2015
407