Pathology Detection and Analysis in Atherosclerosis Images Through Color Clustering Approach
-
Upload
mallisetty-jagadeesh -
Category
Documents
-
view
17 -
download
1
Transcript of Pathology Detection and Analysis in Atherosclerosis Images Through Color Clustering Approach
Pathology Detection and Analysis in Atherosclerosis Images
through Color Clustering Approach
R.Ravindraiah,
Assistant professor, ECE Dept., MITS,
Madanapalle-517325,AP, India, [email protected]
Abstract: Image enhancement and segmentation plays a key role in pathology detection and analysis of medical
images. In this paper we discuss on a most threatening complication of heart patients usually called as Cardio
Vascular Disease (CVD) or Atherosclerosis (also known as arteriosclerotic vascular disease or ASVD). This is a
condition in which an artery wall thickens as a result of the accumulation of fatty materials such as cholesterol
which leads to luminal narrowing and cause Ischemic injury
In this paper we discuss a color clustering algorithm based on hierarchical binary tree region growing approach. Image display devices may allow only a limited number of colors to be simultaneously displayed. This set of
available colors, called a color palette, may be selected by a user from a wide variety of available colors. Such
device limitations make it difficult to represent high definition images which must then be quantized by a palette
with limited size. This color quantization problem is considered in two parts: the selection of an optimal color
palette and the optimal mapping of each pixel of the image to a color from the palette.
Tree structured color palettes greatly reduce the computational requirements of the palette design and pixel mapping
tasks, while allowing colors to be properly allocated to densely populated areas of the color space and hence useful
to cluster different regions in electron micrograph images easily and make the analysis of the pathology easier to a cardiologist.
Keywords: Atherosclerosis disease, Color quantization
I. INTRODUCTION:
Cardiovascular disease is responsible for 80%of deaths among diabetic patients much of which has been
attributed to CAD (coronary artery disease). However, there is an increasing recognition that diabetic patients suffer
from an additional cardiac insult termed „diabetic Cardiomyopathy‟ [1]. The underlying problem in most of the diabetic patients with
CVD is Atherosclerosis leading to narrowing of blood vessels supplying the heart; there by causing heart failure
[2][3]. Atherosclerosis (also known as arteriosclerotic vascular disease or ASVD) is a condition in which an artery
wall thickens as the result of a build-up of fatty materials such as cholesterol. Epidemiological and Clinical trial data
have confirmed the greater incidence and prevalence of heart failure in diabetes [4].
The biological vision system is one of the most important means of exploration of the world to humans,
performing complex tasks with great ease such as analysis, interpretation, recognition and pattern
classification[5].The ultimate aim in a large number of image processing applications is to extract important features from the image data, from which a description, interpretation, or understanding of the scene can be obtained for
human viewers, or to provide `better' input for other automated image processing techniques[6]. Segmentation of
atherosclerosis images (Electron micrograph Images) subdivides into its constituent regions or objects. The level of
detail to which the sub division is carried depends on the problem being solved. Considerable care should be taken
to improve the probability of accurate segmentation. Segmentation algorithms for monochrome images generally are
based on one of two basic categories dealing with properties of intensity values: discontinuity and similarity. In the
first category, the assumption is that boundaries of regions are sufficiently different from each other and from the
background to allow boundary detection based on local discontinuities in intensity. Edge-based segmentation is the principal approach used in this category. Region-based segmentation approaches in the second category are based on
partitioning an image into regions that are similar according to predefined criteria [7].
II. HIERARCHICAL BINARY TREE COLOR CLUSTERING ALGORITHM:
A.COLOR PALETTE DESIGN
The image is assumed to be on a rectangular grid of N pixels. The set of all grid points is denoted by S, and its
members s Є S may be explicitly written as s = (i, j ) where i is the row index and j is the column index. The color
value of the pixel at grid point s is denoted x, = [r,, g,, b,]' where the components are the red, green, and blue tri
stimulus values for the pixel [8] and superscript t means transpose. For a typical monitor, there is a nonlinear
relationship between the input value of the primary and its displayed intensity. For a particular primary such as red,
this relationship may be approximated by
Ir=r
γ
where Z, is the intensity, r is the input to the display, and ϒ= 2.3. More accurate models for this relationship
generally include constant, linear, and logarithmic terms [4]. In order to compensate for this nonlinear relationship
between input and luminance, it is common for video cameras and other image capture devices to pre distort their
output by the inverse relationship:
r= Ir(1/γ)
We shall refer to the pre distorted primary color values as the gamma corrected colors. In effect, this is the
coordinate system being used if one manipulates color image data and directly displays it without correction. It is
important to emphasize that the gamma corrected colors are not linear in intensity, and that the gamma correction,
and therefore the resulting coordinate system, varies with changes in a display's parameters. However, the gamma
corrected colors have the advantage that they may be directly displayed without transformation, and that they are
less susceptible to noise than the corresponding linear primary intensities [ 8].
Alternatively, standard coordinate systems have been defined for the representation of colors. Two such coordinate
systems are the L*a*b* and the L*u*v* color spaces [ 9]. These coordinate systems are nonlinearly related to the
primary intensities, and are chosen so that a fixed Euclidean distance represents a fixed perceptual distance
independent of position in the space. Gentile et al. have studied the application of the L*u*v* coordinate system to
the problems of image quantization [10], halftoning [11], and gamut mismatch [12]. This work indicates that these
coordinate systems can substantially improve quantized image quality relative to standard coordinate systems which
are linearly related to primary luminances. Therefore, it is reasonable to expect that perceptually based coordinate systems such as L*u*v* could improve the quality of displayed images relative to gamma corrected coordinates.
However, it will be assumed throughout this paper that the original image is specified in terms of gamma corrected
coordinates, since this avoids the computational requirements of coordinate transformation methods. All the
techniques developed can, of course, be applied using perceptually based color spaces.
B. BINARY TREE PLATELETE DESIGN
The objective of the palette design algorithm is to partition S into M disjoint sets or clusters where M is the
predetermined palette size determined by hardware constraints. The algorithm proposed here constrains the partitioning of S to have the structure of a binary tree. Each node of the tree represents a subset of S, and the
children of any node partition the members of the parent node into two sets. The set of image pixels corresponding
to node n is denoted Cn. For purposes of illustrating the operations of the algorithm, we will number the nodes so
that the root is 1 and the children of node n are 2n and 2n +1
Since the leaves of the tree form a partition of S, the pixels in S may be coded by representing all pixels in each leaf
node with a single color in the palette. Only the members of the tree's leaf nodes need be stored when the algorithm
is implemented. Each of the leaf node sets can be stored as a linked list. When a node is split, the linked list may be
reorganized into two lists without relocating any of the image color data. The method for generating the binary tree is specified by the number of leaves, M, the method of splitting a node into its two children, and the order in which
the nodes are split. The methods used for splitting nodes and determining which nodes to split both attempt to
minimize the total squared difference between the actual and quantized image.
The TSE is defined by
where qn is the quantization value used to represent the colors in the set Cn. In order to restrict the complexity of the
algorithm, second-order statistical properties will be used to determine the order and manner in which nodes will be
split. The three required cluster statistics are
……(1)
where Rn is a 3 X 3 matrix, and m, is a 3 component vector. Since the mean value of a cluster is the point from
which there is minimum squared deviation, the quantization value of a cluster q, is assumed equal to the cluster mean:
We also define the cluster covariance as
The splitting of a node into two nodes is equivalent to choosing two new quantization levels, q2, q2n+1 and
associating each member of the cluster with the closer quantization level. This, in turn, is equivalent to selecting a
plane which best splits the cluster's colors. In the proposed algorithm, we determine the direction in which the
cluster variation is greatest, and then split the cluster with a plane which is perpendicular to that direction and passes
through the cluster mean. For a large cluster with Gaussian distribution it can be shown that this strategy is optimal.
More specifically, we determine that unit vector e which maximizes the expression
Since Ŕ is symmetric, the solution is the eigenvector e, corresponding to the largest or principal eigenvector λn, of
8. The total squared variation in the direction e, is therefore
……………..(2)
Once the principal eigenvector has been determined, points in C, can be sorted into two sets C2n, and C2n+I in the
following way:
……………….(3)
Finally, the new statistics for each node may be calculated by first calculating R2n, m2n,, and N2n, for node 2n, and
then applying the relations
………….(4)
The order in which nodes are split is chosen with the objective of producing the greatest reduction in TSE at each
stage of the algorithm. Because this strategy does not look ahead to the results of further splits, it is not necessarily
optimal; however, one would expect it to perform well in most practical situations. When a node is split its variance
is mainly reduced along the direction of the principal eigenvector. Therefore, it is reasonable to assume that the reduction in TSE should be proportional to the total squared variation along the direction of the principal
eigenvector e, of R,. For this reason, the principal eigen value X, is used as a measure of the expected reduction in
TSE if node n is split. Given this approximation, the best allocation of a single quantization level is to split the leaf
with the largest principal eigenvector.
The basic binary quantization algorithm may now be described.
1) Set C1 = S.
2) Calculate R 1 , m1, and N1.
3) Do the following M - 1 times: 3.1) Find the leaf n such that X, is largest.
3.2) Use (3) to form the new nodes 2n and 2n+1
3.3) Calculate R, m, and N for new nodes using (4)
RESULTS:
In pathology, an atheroma (“tumor full gruel-like matter”) is an accumulation and swelling in artery walls that is
made up of (mostly) macrophage cells, or debris, that contain lipids (cholesterol and fatty acids), calcium and a variable amount of fibrous connective tissue. Atheroma occurs in atherosclerosis, which is one of the three subtypes
of arteriosclerosis; atherosclerosis, Monckeberg's arteriosclerosis and arteriolosclerosis.
In the context of heart or artery matters, atheromata are commonly referred to as atheromatous plaques. It is an
unhealthy condition, but is found in most humans.
The accumulation (swelling) is always between the endothelium lining and the smooth muscle wall central region
(media) of the arterial tube. While the early stages, based on gross appearance, have traditionally been termed fatty
streaks by pathologists, they are not composed of fat cells, i.e. adipose cells, but of accumulations of white blood
cells, especially macrophages that have taken up oxidized low-density lipoprotein (LDL). After they accumulate large amounts of cytoplasmic membranes (with associated high cholesterol content) they are called foam cells.
When foam cells die, their contents are released, which attracts more macrophages and creates an extracellular lipid
core near the center to inner surface of each atherosclerotic plaque. Conversely, the outer, older portions of the
plaque become more calcific, less metabolically active and more physically stiff over time.
Fig a1: Typical micrograph image of atherosclerosis
Fig a2: the lumen remained is observed
Fig a3: the rupture on the occluded intima is seen
The thick red- thrombus is detected
Variation between Media and plaque is observed
Atherosclerosis plaque [Lipid core]is detected
Fig: a4
The plaque occluded is detected
Fig: a5
The percentage of lumen left is observed here
The gap between media and occluded intima is observed
Calcium content is detected
Fig b1: Histological features of atherosclerosis in humans. A Masson's trichrome stained, coronal section of the left
anterior descending artery of a thrombotically occluded atherosclerotic lesion which led to a fatal myocardial
infarction. Blue stain indicates collagen-rich regions, pale red indicates normal cellular regions, and deep red
indicates the mural thrombus in the lumen of the artery.
Fig b2: clustered image 1
Thrombus is clearly visible
Cholesterol rich core is segmented
Lumen spited and the spread of fibrous cap, thrombus and variation in media and fibrous cap is seen
Ruptured lesion that supports the flow of blood into the plaque is observed
Fig b3:
The accumulated cholesterol rich region is seen.
The extreme borders of the lumen [before and after occlusion] left is detected
We can calculate the area of original lumen and the occluded lumen.
Swelling of the media is observed
Restonic lesion is detected
Fig b4:
The remaining lumen left is detected
Fig b5:
Thrombus is completely detected
Ruptures are seen clearly
Media is detected
Fig b6:
Thick blue- fibrous cap is detected and the ruptures are observed
Gap between the actual and occluded lumen is observed
Percentage of lumen left is observed
The accumulated cholesterol is detected
The swelling in the media is observed
CONCLUSION
Thus the results attained through this method are quite optimun for the analysis of Cardiovascular complications
easily not only for a Cardiologist but even for a novice.
REFERENCES
[1]. “DIABETIC CARDIOMYOPATHY: MECHANISMS, DIAGNOSIS, AND TREATMENT” by Sajad .A.
Hayath, Billal patel, Department of cardiology Northwick Hospital UK, clinlical science (2004).
[2] Fahimuddin.Shaik, Dr.M.N.Giri Prasad, Dr.Jayabhaskar rao,B.Abdul rahim, A.SomaSekhar, “Medical Image
Analysis of Electron Micrographs in Diabetic Patients Using Contrast Enhancement” proceedings of IEEE
International Conference on Mechanical and Electrical technology(ICMET 2010),pp 482-485,Singapore, Sep‟ 10-
12, 2010. [3] Fahimuddin.Shaik,Dr.M.N.Giriprasad, C.Swathi, A.Soma Sekhar, “Detection Of Cardiac Complications In
Diabetic Patients Using Clahe Method” Proceedings Of International Conference On Aerospace Electronics,
Communications And Instrumentation(Aseci-2010), Pp 344-347.6-7 ,India Jan 2010.
[4]. “DIABETIC CARDIOMYOPATHY” by Omar Asghar, Ahmed AL Sunni Sarah withers, the Manchester heart
centre,UK, clinical science(2009).
[5] F.A.Peres, F.R.Oliveira, L.A.neves, M.F.Godoy “Automatic Segmentation of Digital Images Applied in Cardiac
Medical Images” IEEE -PACHE, Conference, Workshops, and Exhibits Cooperation, Lima, PERU, March 15-19,
2010. [6]. DIGITAL IMAGE PROCESSING, by Rafael C. Gonzalez, Richard E.woods -ADDISON-WESLEY. An
imprint of Pearson Education, 1st edition
[7] Tirupati Goskula, Prof. Prasanna .M. Palsodkar ” Extracting Salient Region by Image Segmentation of Color
Images Using Soft Computing Technique” proceedings of International Conference on Emerging Trends in Signal
Processing and VLSI Design (SPVL-2010),pp 515-519, INDIA, June 2010.
[8]. A. Netravali and B. Haskell, Digital Pictures. New York: Plenum, 1988.
[9]. A. Robertson, “The CIE 1976 color-difference formulae,” Color Res. Appl., vol. 2, no. 1, pp. 7-11, Spring 1977.
[10]. R. Gentile, J . Allebach, and E. Walowit, “Quantization of color images based on uniform color spaces,” J.
Imuging Techn., vol. 16, no. 1, pp. 11-21. Feb. 1990.
[11]. R. Gentile, E. Walowit, and J. Allebach. “Quantization and multilevel halftoning of color images for near-
original image quality,” J . Opt. Soc. Amer. A, vol. 7, no. 6, pp. 1019-1026, June 1990.
[12]. R. Gentile, E. Walowit, and J. Allebach, “A comparison of techniques for color gamut mismatch compensation,” J . lmuging Techno. , to be published.