Send Orders for Reprints to reprints@benthamscience.ae
Current Medical Imaging Reviews, 2015, 11, 3-14 3
A Survey on Medical Image Segmentation
Saleha Masood*, Muhammad Sharif, Afifa Masood, Mussarat Yasmin and Mudassar Raza
Department of Computer Science, COMSATS Institute of Information Technology, Wah Cantt,
Pakistan
Abstract: Much work has been done in the field of Image segmentation but still there is a room for
improvement. Medical image segmentation is a sub field of image segmentation in digital image proc-
essing that has many important applications in the prospect of medical image analysis and diagnostics.
Here in this paper different approaches of medical image segmentation will be classified along with
their sub fields and sub methods. Recent techniques proposed in each category will also be discussed
followed by a comparison of these methods.
Keywords: Atlas guided methods, Bayesian method, classifiers, clustering, deformable models, Markov random field, Medical
image segmentation, modalities, neural networks, region growing, thresholding.
1. INTRODUCTION
Segmentation is a process in which an image is divided
into several sub regions based on a specific feature in order
to pick up a region of interest. Segmentation process has
enormous applications in the medical field. In the field of
research and development much work has been done to
overcome the problems faced by the segmentation process
and yet there is a need of more effective and efficient work.
1.1. Purpose of Medical Image Segmentation
In the process of segmentation of a medical image, the
details required by the segmentation process are highly de-
pendent on clinical application of the problem [1]. The pur-
pose of segmentation is to improve the process of visualiza-
tion to handle the detection process more effectively and
efficiently. Other reasons of medical image segmentation can
be seen in Fig. (1).
Fig. (1). Purpose of segmentation.
*Address correspondence to this author at the Department of Computer
Science, COMSATS Institute of Information Technology, Wah Cantt,
Pakistan; E-mail: salehamasood08@gmail.com
The analysis of functions of anatomy problems is carried
out through the segmentation process [2]. It covers all the
factors that influence the analysis of a disease. Through the
process of segmentation one can analyze, diagnose, quantify,
monitor and plan the navigation of a disease.
1.2. Basic Principles of Segmentation
The process of segmentation is carried out on the basis of
two central principles. These principles as shown in Fig. (2)
are classified on the basis of features that contain texture,
intensity, sharpness of edges and all the significant features
in this context [3].
Fig. (2). Basic principles of segmentation.
1.3. Problems in Medical Image Segmentation
Segmentation of medical image faces many problems be-
cause of which the quality of segmentation process gets af-
fected [4]. These problems can be analyzed in Fig. (3) below.
The problem of uncertainty arises when there is noise in
the image which makes the classification of image difficult
[5]. The reason is that intensity values of pixels are amended
because of noise in the image. This alteration in the intensity
values of pixels disturbs uniformity in the intensity range of
1875-6603/15 $58.00+.00 ©2015 Bentham Science Publishers