A survey on content based medical image retrieval for

IJRET: International Journal of Research in Engineering and Technology eISSN: 2319-1163 | pISSN: 2321-7308
__________________________________________________________________________________________
Volume: 03 Special Issue: 07 | May-2014, Available @ http://www.ijret.org 149
A SURVEY ON CONTENT BASED MEDICAL IMAGE RETRIEVAL FOR
MRI BRAIN IMAGES
G.Sharmila1
, D.Saraswathi2
, R.Ramkumar3
1
PG Scholar, ECE, Manakula Vinayagar Institute of Technology, Puducherry, India
2
Assistant Professor, ECE, Manakula Vinayagar Institute of Technology, Puducherry, India
3
PG Scholar, ECE, Manakula Vinayagar Institute of Technology, Puducherry, India
Abstract
BRAIN tumor is also a common curable disease, when it is diagnosed at the earlier stage. Diagnose of disease is depend on how
efficiently the query image abnormality is accurately detected. The computer aided automated system such as content based medical
image retrieval technique is used to retrieve query based images in the large database using combination of feature extraction and
similarity matching methods. In this paper, we present an overview of CBMIR technique to classify brain images. This paper also
includes literature survey on feature extraction and feature selection method with similarity matching algorithm. The main objective
of this paper is to give comparison on existing methods and highlight the best method to retrieve similar tumor image from a large
database.
Keywords: Content Based Medical Image Retrieval (CBMIR), feature extraction, feature selection, similarity matching,
magnetic resonance images (MRI), brain tumor images.
-----------------------------------------------------------------------***----------------------------------------------------------------------
1. INTRODUCTION
Content based image retrieval plays an important role in many
applications especially in the medical field. Because of
thousands of images produced by the hospitals everyday
labeling the image and classifying abnormality by human
intervention is a time consuming and difficult task. Further
treatment planning is depend on how accurately the
abnormality is detected , false detection may also cause wrong
diagnosis leads to serious problem to face this situation the
best content based image retrieval method is used to retrieve
query based image from large database .
Accurate result can be produced only with the help of
automated computer aided technique. CBMIR can be useful
for many diseases such as brain tumor, breast cancer, spine
disorder problem etc which is acquired through many
modalities such as CT scan, MRI, mammogram etc. In this
paper we concentrate on MRI brain tumor images because
brain is an important sense organ which controls and
coordinate the whole homeostatic body function include heart
beat blood circulation etc. So the accurate classification of
brain tumor images is an important part of query based image
retrieval from a large database. Retrieval of similar images is
based on similarity matching measures between features of
query image with the database images features. The first step
in this method is preprocessing because noise or misalignment
will occur usually while capturing the images. After that
feature extraction takes place such as content of image is
represented using set of feature to avoid large input to the
processing device and also we can‘t assure that all the content
feature does not contain the useful information so the
redundant features are reduced using some of feature selection
method. Hopefully, this survey will help the medical
Practitioner to detect the disease accurately for further
treatment and also gives idea for innovation of hybrid
technique for research scholar those who work in this area.
This paper gives an overview of available literature in the field
of content-based access to medical image data and on the
technologies used in the field. Section II provides literature
survey on image pre-processing. Section III contains the
details of extracted features from an image. Section IV
presents the literature survey on feature extraction techniques.
Section V presents the literature survey on feature selection
techniques. Section VI gives a comparative analysis on
similarity matching and classification techniques. Section VII
summarizes several concluding remarks and future work.
Most of these systems have a very similar architecture for
browsing and archiving/indexing images comprising tools for
the extraction of visual features, for the storage and efficient
retrieval of these features, for distance measurements or
similarity calculation and a type of graphical user interface
(GUI). This general system setup is shown in Fig. 1.

__________________________________________________________________________________________
2. LITERATURE SURVEY ON IMAGE PRE-
PROCESSING
Image preprocessing is the first step of image retrieval to
ensure accuracy of subsequent steps. The images acquired
through different modalities cause many artifacts such as low
resolution, noise and extra cranial tissue etc. which reduces the
accuracy of acquired result.
In order to overcome the above problem preprocessing of an
image is required. An analysis on filtering techniques such as
Gabor & QMF filters for noise is performed by [1].These
primitive methods along with reducing the noise blur the
important and detailed structure necessary for subsequent
steps. In order to increase the processing speed and to reduce
the error probability of mammogram images Morphological
top hat filtering algorithm is utilized in [2]. But these types of
filtering are applicable for mammogram images only. To
eliminate the noise in images, Gaussian filter is suggested in
[3].
Fig -1: A framework of Content Based Medical Image
Retrieval System
The advantage of this filter is to reduce the noise as well to
increase the contrast and intensity of an image. To improve
image quality histogram equalization, edge detection, noise
filtering and thresholding is proposed in [4]. Diffusion filtering
combined with simple non-adaptive intensity thresholding is
used to enhance the region of interest in [5]. The main
drawback of this technique is non -adaptive nature of the
threshold value. The noise removal technique using wavelet
and curve let is implemented in [6]. Hybrid approach
involving variance stabilizing transform (VST) are also used.
But this technique is applicable only for image with Poisson
noise. Combination of approaches such as local maximum,
local minimum, local (max-min)/2, local mean, local median,
local Mod are given in [7]. This will improve the image
information content by suppressing the undesired distortions.
A contras agent accumulation model is implemented in [8].
This improves only the contrast characteristics of an image but
it is not eliminating the unwanted tissue. Weiner filter is used
to remove noise in MR brain image is used in [9]. Apart from
noise removal several other preprocessing steps are also to be
considered in CBMIR. This includes image format conversion,
image type conversion etc. the combination of three modalities
of MR images for further processing is proposed in [10].
Though several preprocessing method are mentioned for
removal of noise in an image, the removal of unwanted tissue
and preserving needed image information content without
blurring is highly essential for accurate retrieval of image from
a large database. Some of the Pre-processing techniques are
tabulated below:
Table -1 Comparison of various preprocessing methods
METHO
DS
CHARACTERISTICS DISADVANTAGE
S
Wavelet
Transfor
m [6]
Analysis image in
different frequencies
Missing of some
original information
during
decomposition.
Threshold
ing [4]
Enhance the region of
interest
Non-Adaptive
nature of threshold
value
Gabor
and QMF
[1]
Removes only the noise Blur the details
present in an image
Morpholo
gical
operator
Used to segment the
region of interest
Doesn‘t remove
noise or improve
image quality
3. TYPES OF FEATURES
3.1 Visual Features
This section describes the visual features that are used in the
various applications. Feature contains information about image
content that is needed to retrieve similar query images from
database and also to classify the type of disease in the medical
field. Visual features were classified in [12] into primitive
features such as color or shape, logical features such as
identity of objects shown and abstract features such as
significance of scenes depicted. Still, all currently available
systems only use primitive features unless manual annotation
is coupled with the visual features as in [13].
QUERY IMAGE
IMAGE PRE-PROCESSING
FEATURE EXTRACTION
FEATURE SELECTION
SIMILARITY MATCHING
RETRIEVED
IMAGES
MEDICAL IMAGE
DATABASE

__________________________________________________________________________________________
3.2 Color Features
In stock photography (large, varied databases for being used
by artists, advertisers and journalists), color has been the most
effective feature and almost all systems employ colors.
Although most of the images are in the red, green, blue (RGB)
color space, this space is only rarely used for indexing and
querying as it does not correspond well to the human color
perception. Much effort has also been spent on creating color
spaces that are optimal with respect to lighting conditions or
that are invariant to shades and other influences such as
viewing position [14, 15]. This allows identifying the colors
even under varying conditions but on the other hand
information about the absolute colors is lost. But in the
medical domain, absolute color or grey level features are often
of very limited expressive power unless exact reference points
exist as it is the case for computed tomography images.
3.3 Shape Features
Shape describes the edge of an image however result of
segmentation is also a one of the shape feature. Pattern
recognition based on color and shape is mentioned in [22].
Fast and effective retrieval of tumor shape is given in [23].
For patients with poor MRI quality, the texture and intensity
features may prove inadequate for PF tumor segmentation.
Though it gives information about images, the internal tissue
present in a tumor image may lost and it causes
misclassification.
3.4 Texture Features
Due to complex tissues present in an MR brain images
classification through shape feature is difficult. For accurate
classification, features that are used in medical field is called
texture. Texture is a commonly used feature in the analysis
and interpretation of images. It is characterized by a set of
local statistical properties of pixel intensities. Analysis of
texture feature is mentioned in [7] which improve the accuracy
of classification.
3.5 Features used in Medical Field
As color and grey level features are of less importance in
medical images than in stock photography, the texture and
shape features gain more importance in [26]. Basically all of
the standard techniques for texture characterization are used
from edge detection using Canny operators [16] to Sobel
descriptors [17]. [18,21,19] also use Fourier descriptors to
characterize shapes, [16,20,19] use invariant moments and
[16] also scale-space filtering.
4. LITERATURE SURVEY ON FEATURE
EXTRACTION
The intention of Feature extraction is to reduce the original
data set by measuring certain properties, or features, that
distinguish one input pattern from another pattern [25]. The
extraction of feature vector which consists of various feature
components. It is generated to represent the content of each
image in the database with accuracy and uniqueness. Spatial
gray level co- occurrence matrix estimate the image properties
related to second order statics. Haarlick [32] suggested the use
of GLCM is one of the most well known method. The wavelet
based texture feature for classification is used by [28]. Multi-
fractional Brownian motion (MBM) algorithm is used in [27].
The advantage of this method is image with different
resolution gives same result. Modified Haar wavelet
transformation is proposed in [29]. The texture features
namely Contrast, correlation Homogeneity and Energy is used
in [30]. This improves specificity and accuracy of retrieved
image. Auto color correlogram and correlation in [31] will get
accuracy in less iteration. But the iteration is depending on the
need of application. EI-Sayed et al. [24] has obtained the
features related with MR images using discrete wavelet
transformation (DWT). But all the techniques which use basic
discrete wavelet transform which does not yield superior
result. An improved version based on wavelet packet
decomposition is implemented in [32]. The results revealed
that the extracted features are efficient than previous method.
From the above mentioned techniques, it is clear that the
extraction of appropriate features will improve the accuracy
for classification and similarity matching.
Table- 2 Mean square Error and classification accuracy of
different feature extraction technique
Stages
Mean Square Error
Classification
Accuracy%
GL
CM
HAR
ALIC
K‘S
WSF
TA
GLC
M
HAR
ALIC
K‘S
WSF
TA
Trainin
g
1.59
6
2.235
0.058
5
98.6
%
89%
100
%
validati
on
2.34
5
8.632 1.203
98.7
%
87.6%
100
%
Testing
1.89
5
2.215
4
1.003
96.4
%
88.9%
98.0
%

__________________________________________________________________________________________
Chart -1: performance of different feature extraction
technique
5. LITERATURE SURVEY ON FEATURE
SELECTION
Feature selection is the process of reducing the dimension of
feature vectors. For feature selection, every feature is
observed. Significant features are selected by the calculation
of mean values for every feature in benign tumour class and
malignant tumour class. Student‘s t-test [34] is employed as a
method to separate both the classes. Principal component
based analysis is used in [33] gives better feature subset, but
fresh training is needed whenever it encounter a new images.
The ability of genetic algorithm for pattern classification is
explored in [34]. The samples belonging to the same class is
only accepted by GE. Reduction of features using Linear
Discriminant Analysis is proposed in [35].
An another evolutionary approach namely, Partial swam
intelligence is used by [36]. This method is used for
optimizing the weight of Artificial Neural Network for image
classification application. Since PSO involves many random
parameters, the weight obtained through this may not be a
stabilized set of weight. Ant Colony based feature selection is
given in [37]. By selecting the features in terms of feature set
size , its computation increases when feature size increases.
Optimal Fuzzy rule selection for classification is implemented
in [38]. But this technique is highly sensitive to change in the
parameters of the membership functions. The hybrid of above
techniques will give even better results for eg: (wavelet+
Principal component analysis) or (genetic algorithm+ fuzzy
rule) etc is mentioned in [39]. But time complexity is more in
the hybridized method which makes system non- feasible. The
above mentioned feature selection methods can be used for
calculating classification accuracy and similarity matching.
6. LITERATURE SURVEY ON CLASSIFICATION
Automatic similarity matching is used to retrieve similar
images from large database that reduces the cost of
classification work of human. One of the primary distance
measure techniques is Euclidean distance which is used in
[39]. The dimensionality is often reduced to work efficiently
with these distances even in large database. The neural
network is used for classification of mammogram images [40].
Other statistical approach Bayesian network is used in [42].
Classification using Support Vector Machine is mentioned in
[41] for MR brain images which is the simplest one and gives
better accuracy. The Hybrid technique of SVM and
probabilistic neural network (PNN) classifiers is proposed in
[34]. The above mentioned techniques gives an improved
accuracy for classification and image retrieval.
7. CONCLUSIONS AND FUTURE WORK
We have categorically presented a detailed survey of the
prominent methods of automated technique for medical image
retrieval published in the literature. In the case of Pre-
processing stage, various types of filter techniques have been
discussed. In the case of feature extraction and feature
selection some of the evolutionary techniques used in the
literature have also been presented. The performance depends
on the appropriate extracted features and the features that are
selected for similarity matching [46]. We hence presented a
survey on all stages as a whole to implement the improvised
CBMIR system, which has a rich potential and scope of
application in medical domain. CBMIR is an ever- growing
rich field, with a wide scope of interdisciplinary research over
the fields of Biomedical, Computer Science, Mathematics,
Statistics and Machine learning.
Table-3: Classification accuracy of different metrics
TECHNIQUES
CLASSIFICATION
ACCURACY
Minkowski 79.40
Chebyshev 79.20
L2 79.80
Standardized L2 79.20
Mahalanobis distance 80.80
City block 81.40
Correlation 81.60
Relative deviation 81.00
L1 81.60
cosine 82.00
Spearman 82.40
0
20
40
60
80
100
120
MEAN SQUARE
ERROR
CLASSIFICATION
ACCURACY %

__________________________________________________________________________________________
In the future, one can expect rich concepts from widely
varying areas such as information theory, optimization theory,
rough-fuzzy sets, soft computing and evolutionary techniques
etc., to contribute and enrich this problem domain in the field
of pattern recognition.
REFERENCES
[1] Sebe N, Michel S.L ―Wavelet based texture
classification‖. 15th
Int. conf. on pattern recognition
2000, Vol 3: 947-50.
[2] D. Cascio, F. Fauci, R. Magro, G. Raso,―Mammogram
Segmentation by Contour Searching and Mass Lesions
Classification With Neural Network‖ IEEE
Transactions On Nuclear Science, Vol. 53, No.
5,October 2006.
[3] Nitish Zulpe and Vrushsen Pawar, ―GLCM Textural
Features for Brain Tumor Classification‖ International
Journal of Computer Science Issues, Vol. 9, Issue 3, No
3, May 2012.
[4] Ms.suchithra swami, ―Brain tumor detection using
unsupervised learning based neural network‖ Int. conf
on comm system and network technologies, IEEE
2013.
[5] Yang Y, Hung S, ―Novel statistical approach for
segmenting of brain magnetic resonance imaging using
an improved expectation maximization algorithm‖
Optica Applicata 2006 ;36:125-36.
[6] Zhang B, Jalal M, Strack J. ― Wavelet , Ridgelets and
curvelets for poisson noise removal‖ IEEE Trans.Image
.proc.2008;17:1093-1108.
[7] B.V. Ramana Reddy et.al ― Classification of Textures
Based on Features Extracted from Preprocessing
Images on Random Windows‖ Int. journal of
Advamced Science and technology 2009; vol 9.
[8] Prastawa M, bullit E, Gerig G. ―simulation of brain
tumors in MR images for evalution of segmentation
efficiency‖ Medical image analysis 2009; 13: 297-311.
[9] Ratan R, Sharma s, Sharma s. ― Multiparameter
segmentation and quantization of brain tumor from MR
images‖. Int. jrnl of science and technology 2009; 2:
11-5.
[10] Ratan R, Sharma S ,Sharma S. ―Brain tumor detection
based on multi-parameter MR image analysis‖.
Graphics, vision and image processing Journal 2009; 9:
9-17.
[11] D. Judehemanth, C. keziselvavijila, J. anitha ‗Artifical
intelligent based Brain Pathology Identification
techniques in Magnetic resonance Images‖. Int.jrnl of
review in computing. 2009.
[12] J.P. Eakins, M.E. Graham, ―Content-based image
retrieval‖ ,Tech. Rep. JTAP-039, JISC Technology
Application Program Newcastle upon Tyne, 2000.
[13] T. Pfund, S. Marchand-Maillet, ―Dynamic multimedia
annotation tool‖ Proceedings of the SPIE Photonics
West Conference on Internet Imaging III, vol. 4672,
San Jose, CA, USA, 2002, pp. 216—224
[14] J.-M. Geusebroek, R. van den Boogaard, A.W.M.
Smeulders, H. Geerts, ―Color invariance‖ , IEEE Trans.
PatternAnal. Machine Intel.23 (12) (2001) 1338—
1350.
[15] M. Ortega, Y. Rui, K. Chakrabarti, K. Porkaew, S.
Mehrotra,T.S. Huang, ―Supporting ranked boolean
similarity queries in MARS‖, IEEE Trans. Knowledge
Data Eng. 10 (6) (1998) 905—925.
[16] K. Veropoulos, C. Campbell, G. Learnmonth, ―Image
processing and neural computing used in the diagnosis
of tuberculosis‖, Proceedings of the Colloquium on
Intelligent Methods in Healthcare and Medical
Applications (IMHMA), York, UK, 1998.
[17] C. Brodley, A. Kak, C. Shyu, J. Dy, L. Broderick, A.M.
Aisen, ―Content-based retrieval from medical image
databases: A synergy of human interaction, machine
learning and computer vision‖ ,Proceedings of the 10th
National Conference on Artificial Intelligence,
Orlando, FL, USA, 1999, pp. 760—767.
[18] S. Antani, L.R. Long, G.R. Thoma, ―A biomedical
information system for combined content-based
retrieval of spine X-ray images and associated text
information‖, Proceedings of the Third Indian
Conference on Computer Vision, Graphics and Image
Processing (ICVGIP 2002), Ahamdabad, India, 2002.
[19] J.M. Bueno, F. Chino, A.J.M. Traina, C.J. Traina, P.M.
Azevedo-Marques, ―How to add content-based image
retrieval capacity into a PACS‖, Proceedings of the
IEEE Symposium on Computer-Based Medical
Systems (CBMS 2002), Maribor, Slovenia, 2002, pp.
321—326.
[20] M.E. Mattie, L. Staib, E. Stratmann, H.D. Tagare, J.
Duncan, P.L. Miller, PathMaster ― Content-based cell
image retrieval using automated feature extraction‖, J.
Am. Med Informatics Assoc. 7 (2000) 404—415.
[21] A.K. Jain, A. Vailaya, ―Image retrieval using color and
shape‖, Pattern Recog. 29 (8) (1996) 1233—1244.
[22] P. Korn, N. Sidiropoulos, C. Faloutsos, E. Siegel, Z.
Protopapas, ―Fast and effective retrieval of medical
tumor shapes‖, IEEE Trans. Knowledge Data Eng. 10
(6) (1998) 889—904.
[23] E.-S. A EI-Dahshan, T. Hosny, and A-B. M. Salem,
"Hybrid intelligent techniques for MRI brain images
classification," Digital Signal Processing, vol. 20, no. 2,
pp. 433-441, 2010.
[24] Lawrence O Hall, ― A comparison of neural network
and Fuzzy clustering techniques in segmenting
magnetic resonance images of the brain‖, IEEE Trans
on neural networks, Vol 3, No.5,sep 1992.
[25] Ani1. k. jain, ―Fundamentals of digital image
processing‖ prentice-hall publication, ISBN: 0-13-
336165-9, 1989.
[26] Henning Müller, ―A review of content-based image
retrieval systems in medical applications–—clinical

__________________________________________________________________________________________
benefits and future directions‘ International Journal of
Medical Informatics (2004) 73, 1-23
[27] Atiqislam, syed M.S reza and k.m. iftekharuddin ―Multi
fractal texture estimation for detection and
segmentation of brain tumor” IEEE
Trans.on.medical,2013.
[28] Arivazhagan S, Ganesan L., ― Texture classification
using wavelet transform‖. Pattern recog letters 2003;24:
1513-21.
[29] P.S. Malge et.al ―Performance Evaluation of Texture
based Image Retrieval‖ International Journal of
Computer Applications (0975 – 8887) Volume 72–
No.2, May 2013
[30] RamadassSudhir et. Al, ― A Efficient Content based
Image Retrieval System using GMM and Relevance
Feedback‖, International Journal of Computer
Applications (0975 – 8887) Volume72– No.22, June
2013.
[31] R. M Haarlick, ―Statistical and structural approaches to
texture‖, proceeding of IEEE,vo1.67, pp.786-804,
1979.
[32] Hiremath Pet.al ―Texture classification using wavelet
packet decomposition‖ Graphics, vision, and image
processing journal 2006;6:77-80.
[33] Kishore K,Patnaik M, Mani V, Agarwal K.
―Applications of genetic programming for multi
category parttern classification‖ , IEEE trans.on
Evolutionary computation 2009; 4: 242-58.
[34] A. Padma Nanthagopal, R. Sukanesh, ―Wavelet
statistical texture features-based segmentation and
classification of brain computed tomography images‖
IET Image Processing 2012.
[35] V.P.Gladis Pushpa Rathi and Dr.S.Palani ―Brain
Tumor MRI Image Classification With Feature
Selection And Extraction Using Linear Discriminant
Analysis‖ 2012.
[36] Chandramouli k, Izquierdo E , ―Image classification
using chaotic particle swam optimization‖ IEEE Int.
conf, on image processing 2006; 3001-04.
[37] Ling Chen, Bolun Chen, Yixin Chen, ― Image Feature
Selection Based on Ant Colony Optimization‖ 2010.
[38] Chakraborty D,Nikhil P.A, ―Neuro fuzzy scheme for
simultaneous feature selection and Fuzzy rule based
classification‖ IEEE Trans, neural network 2004; 15:
110-23.
[39] Aditi Mehta , ―Review and Comparison of Various
Feature Extraction Techniques in CBIR‖, International
Journal of Computer Applications (0975 – 8887)
Volume 71– No.23, June 2013.
[40] R. Hanka, T.P. Harte, ―Curse of dimentionality:
classifying large multidimensional images with neural
network‖, Proceedings of the European workshop
Computer – intensive methods in control and signal
processing (CIMCSP 1996), prague, Czech Republic,
1996.
[41] Magadiet.al, ―An intelligent model for automatic brain
tumor diagnosis bared on MR images‖.
[42] Aditi P. Killedar, ― Content Based Image Retrieval
Approach to Tumor Detection in Human Brain Using
Magnetic Resonance Image‖ 1st International
Conference on Recent Trends in Engineering &
Technology, Mar-2012.
[43] Y. Liu, F. Dellaert, ―Classification – driven medical
image retrieval‖, Proceedings of the ARPA image
understaning workshop, 1997.
[44] D.Saraswathi, E.Srinivasan, G.Sharmila, ―An
Automated Diagnosis system using Wavelet based
SFTA Texture Features‖ IEEE Int conf., ICICES, 2014.
[45] D.Saraswathi, Chanemouga priya, G.Sharmila, ―An
Efficient Color Feature Extraction Method for Content
Based Image Retrieval‖, NCWCAPS‖13, pg: no: 7-
11,2013.
[46] D.Saraswathi, E.Srinivasan, G.Sharmila, ―A
Comprehensive study of Classification Techniques
used in Medical Field ‖ ICIEES, Int conf., PSG college
of Technology, PP. 233-238, 2013.
BIOGRAPHIES
Sharmila Govindaraj received her B.E.
degree in Electronics and communication
Engineering from Mailam Engineering
College, Anna University, Chennai in 2011
and currently doing her Master degree in
Electronics and communication engineering
at Manakula Vinayagar Institute of Technology, Pondicherry
University, Puducherry. Her fields of interests are Image
Processing, Digital Signal Processing and Electronic Devices
Saraswathi Velmurugan received her
M.E degree in Applied Electronics from
PSG College of Technology, Anna
university, Coimbatore in 2009 and
currently working as an Assistant
Professor at Manakula Vinayagar Institute
of technology, Pondicherry University, Puducherry. She is
pursuing her Ph.D. from Pondicherry University, Puducherry.
She has ten years long experience in the field of teaching. Her
research interest includes Image Processing, Medical Image
Classification, Image retrieval and Recognition Systems.
Ramkumar Ramachandran received his
B.E. degree in Electronics and
communication Engineering from Mailam
Engineering College, Anna University,
Chennai in 2011 and currently doing his
Master degree in Electronics and
communication engineering at Manakula Vinayagar Institute
of Technology, Pondicherry University, Puducherry. His field
of interests are Image Processing, Wireless and
Telecommunication and Computer Networks.

A survey on content based medical image retrieval for

More Related Content

What's hot (19)

Viewers also liked (20)

Similar to A survey on content based medical image retrieval for (20)

More from eSAT Publishing House (20)

Recently uploaded (20)

A survey on content based medical image retrieval for