brain tumor classification using machine learning algorithms. brain-tumor... · 2018. 7. 31. ·...

P-ISSN: 2347-4408

E-ISSN: 2347-4734

30| Page April 2017, Volume - 4, Issue - 2

Brain Tumor Classification Using Machine Learning Algorithms

B.Balakumar1, P.Raviraj2, E.Divya Devi3

1Assistant Professor, Centre for Information Technology and Engineering, Manonmaniam Sundaranar University, Tirunelveli,

India, [email protected] 2Professor, Department of CSE, Kalaignar Karunanidhi Institute of Technology, Coimbatore, Tamilnadu, India,

[email protected] 3PG Scholar, Centre for Information Technology and Engineering, Manonmaniam Sundaranar University, Tirunelveli, India,

[email protected]

Abstract: A brain tumor is a collection of tissue that is grouped by

a gradual addition of anomalous cells and it is important to classify

brain tumors from the magnetic resonance imaging (MRI) for

treatment. Human investigation is the routine technique for brain

MRI tumor detection and tumors classification. Interpretation of

images is based on organized and explicit classification of brain

MRI and also various techniques have been proposed. Information

identified with anatomical structures and potential abnormal

tissues which are noteworthy to treat are given by brain tumor

segmentation on MRI, the proposed system uses the adaptive pillar

K-means algorithm for successful segmentation and the

classification methodology is done by the two-tier classification

approach. In the proposed system, at first the self-organizing map

neural network trains the features extracted from the discrete

wavelet transform blend wavelets and the resultant filter factors

are consequently trained by the K-nearest neighbour and the

testing process is also accomplished in two stages. The proposed

two-tier classification system classifies the brain tumors in double

training process which gives preferable performance over the

traditional classification method. The classifiers can accurately

classifying the status of the brain image into normal / abnormal.

Keywords: Image segmentation, MRI (Magnetic Resonance

Imaging), Adaptive pillar k- means algorithm

1. INTRODUCTION

Brain tumor is one of the major causes of death among other types of the cancers. Proper and timely diagnosis can prevent

the life of a person to some extent. Therefore we have

proposed an automated reliable system for the diagnosis of

the brain tumor. Proposed system is a two-tier system for

brain tumor diagnosis and tumor region extraction. First,

noise removal is performed as the pre-processing step on the

brain MR images. Texture features are extracted from these

noise free brain MR images. Next phase of the proposed

system is Self-organizing Mapping based feature training and

followed by the SVM based classification that is based on

these extracted features. More than 89.5% accuracy is

achieved by the classification phase. Results of the proposed

technique show that tumor images are recognized quite

accurately. This technique has been tested against the datasets

of different patients received from Brainix dataset where the

proposed work is tested on 30 normal images and 30

abnormal images.

2. LITERATURE SURVEY

The brain tumour identification from a MRI is a complex

process hence artificial intelligence is applied to detect the

exact tumour position in a brain MRI. In this paper we are

developed a novel technique for the tumour segmentation and

classification in brain MRI. The proposed approach for the

brain MRI classification is implemented in the working

platform of Matlab and the detailed explanation on the

implementation performance measure is as follows. Brain

tumour is most treatable and curable if captured at the initial

stages of the infection. This locates exorbitant pressure on the

brain, causing an incremented intracranial pressure and can

cause permanent brain damage and eventually death.

Untreated or progressive brain cancer can only spread inward

because the skull will not let the brain tumour expands

outward. Only invasive techniques such as biopsy and spinal

tap methods can determine whether the brain tumour is

cancerous or non-cancerous. However the system designed in

this work avails segmenting the MR image with Adaptive

Pillar K means clustering algorithm and classifying

cancerous and non-cancerous brain tumors automatically by

the

two- tier classifier approach, which uses the statistical texture

features extracted by DWT. To demonstrate and evaluate the

mailto:[email protected]:[email protected]

P-ISSN: 2347-4408

E-ISSN: 2347-4734


performance of the proposed system, mainly applied to brain

MRI.

The skull captured in the brain MRI is removed by the skull

stripping technique. The segmentation process is done by

using Adaptive Pillar k means technique to optimize the

computation time and hence improved the precision and

enhance the quality of image segmentation. All classification

result could have an error rate and on occasion will either fail

to identify an abnormality, or identify an abnormality which is

not present.

3. METHODOLOGY

3.1 Existing system

Optical flow analysis Drawback: sensitive to noisy environment and it is

not suitable for real time implementations

Temporal difference analysis Drawback: The extracted shapes of moving objects

are generally incomplete, especially when the moving

objects in a scene are stationary or exhibit slow motion.

Background subtraction Drawback: increased time complexity

Bayesian approach Drawback: computational effort is high and not

suitable for long video files.

3.2 Proposed system

The multi-background generation module effectively generates a flexible probabilistic model through an

unsupervised learning process to fulfill the property of

either dynamic background or static background.

The moving object detection module achieves complete and accurate detection of moving objects by only

processing blocks that are highly likely to contain moving

objects

3.3 Advantages

Very efficient Dynamic scenes are very easy to detect Low memory capacity needed Scalable Robust for any kind of video

4. PRE-PROCESSING TECHNIQUES

Brain MRIs are degraded during the process of imaging due

to image transmission and image digitization by noise and

existence of extra-cranial tissues in MRI such as Skull, bone,

skin, air, muscles, and fat.

Pre-processing is a procedure to eliminate these noises and

extra-cranial tissues from the Brain MRI and alters the

heterogeneous image into homogeneous image. Though there

are lots of filters which have been used for filtering the

images, some of them corrupt the miniature details of the

image and some conventional filters will process the image

incessantly (smoothing)and consequently harden the edges of

the image. Hence, the proposed pre-processing steps namely

De-noising and skull stripping provide better Image clarity.

4.1 Sobel Operator

The operator consists of a pair of 3×3 convolution kernels as

shown in Figure 1. One kernel is simply the other rotated by

90°.

These kernels are designed to respond maximally to edges

running vertically and horizontally relative to the pixel grid,

one kernel for each of the two perpendicular orientations. The

kernels can be applied separately to the input image, to

produce separate measurements of the gradient component in

each orientation (call these Gx and Gy). These can then be

combined together to find the absolute magnitude of the

gradient at each point and the orientation of that gradient. The

gradient magnitude is given by:

Typically, an approximate magnitude is computed using:

P-ISSN: 2347-4408

E-ISSN: 2347-4734


which is much faster to compute.

The angle of orientation of the edge (relative to the pixel grid)

giving rise to the spatial gradient is given by:

4.2 Robert’s cross operator:

The Roberts Cross operator performs a simple, quick to

compute, 2-D spatial gradient measurement on an image.

Pixel values at each point in the output represent the estimated

absolute magnitude of the spatial gradient of the input image

at that point.

The operator consists of a pair of 2×2 convolution kernels as

shown in Figure. One kernel is simply the other rotated by

90°. This is very similar to the Sobel operator.

These kernels are designed to respond maximally to edges

running at 45° to the pixel grid, one kernel for each of the two

perpendicular orientations. The kernels can be applied

separately to the input image, to produce separate

measurements of the gradient component in each orientation

(call these Gx and Gy). These can then be combined together

to find the absolute magnitude of the gradient at each point

and the orientation of that gradient. The gradient magnitude is

given by:

although typically, an approximate magnitude is computed

using:

which is much faster to compute.

The angle of orientation of the edge giving rise to the spatial

gradient (relative to the pixel grid orientation) is given by:

4.3 Prewitt’s operator:

Prewitt operator is similar to the Sobel operator and is used

for detecting vertical and horizontal edges in images.

4.4 Laplacian of Gaussian:

The Laplacian is a 2-D isotropic measure of the 2nd spatial

derivative of an image. The Laplacian of an image highlights

regions of rapid intensity change and is therefore often used

for edge detection. The Laplacian is often applied to an image

that has first been smoothed with something approximating a

Gaussian Smoothing filter in order to reduce its sensitivity to

noise. The operator normally takes a single gray level image

as input and produces another gray level image as output.

The Laplacian L(x ,y) of an image with pixel intensity values

I(x ,y) is given by:

Since the input image is represented as a set of discrete pixels,

we have to find a discrete convolution kernel that can

approximate the second derivatives in the definition of the

Laplacian. Three commonly used small kernels are shown in

Figure 1.

http://homepages.inf.ed.ac.uk/rbf/HIPR2/gsmooth.htm

P-ISSN: 2347-4408

E-ISSN: 2347-4734


Figure 1 Three commonly used discrete approximations to

the Laplacian filter.

Because these kernels are approximating a second derivative

measurement on the image, they are very sensitive to noise.

To counter this, the image is often Gaussian Smoothed before

applying the Laplacian filter. This pre-processing step reduces

the high frequency noise components prior to the

differentiation step.

In fact, since the convolution operation is associative, we can

convolve the Gaussian smoothing filter with the Laplacian

filter first of all, and then convolve this hybrid filter with the

image to achieve the required result. Doing things this way

has two advantages:

Since both the Gaussian and the Laplacian kernels are usually much smaller than the image, this method

usually requires far fewer arithmetic operations.

The LOG (`Laplacian of Gaussian') kernel can be pre calculated in advance so only one convolution needs

to be performed at run-time on the image.

Denoising

In the pre-processing of brain MRI, the noise will be removed

by utilizing the non-local mean filter which does not update a

pixel’s value with an average of the pixels around it, instead updates it using a weighted average of the pixels judged to be

most kindred. The weight of each pixel depends on the

distance between its intensity grey level vector and that of the

target pixel.De-noised image of each pixel i of the non-local

means is computed with the following equation

Where, j is the noisy image and N is the de-noised image, and

weights w(i, j) meet the following conditions 0 ≤ w(i, j) ≤ 1.

Each pixel is a weighted average of all the pixels in the image

which is based on the similarity between the neighborhoods

of pixels i and j.

Adaptive pillar K means algorithm with the efficient

distance metric

The system utilizes the authentic size of the image to perform

high quality image segmentation which causes high-resolution

image data points to be clustered. Therefore utilize the K-

means algorithm for clustering image data by considering that

it has ability to cluster immensely colossal data and

additionally outliers’ payments are utilized expeditiously and efficiently. Because of starting points engendered arbitrarily,

one of the local minima leads to erroneous clustering results

hence K-means algorithm is arduous to reach global optimum.

To evade this phenomenon, the proposed system utilizes

adaptive pillar algorithm, which is very robust and superior

for initial clusters optimization for K-means by deploying all

centroids far discretely among them in the data distribution.

This algorithm is inspired by the cerebration process of

determining a set of pillars’ locations to make a stable house or building. In the proposed adaptive pillar K-means

algorithm we modified the pillar K-means algorithm in[26].

In the proposed adaptive pillar K-means algorithm we find

out the average mean of the data point instead of grand mean

in the previous algorithm. The average mean based initial

centroid point selection can improve the performance of the

clustering than the grand mean based existing method. Locate

two, three, and four pillars, to withstand the pressure

distributions of several different roof structures composed of

discrete points. It is inspiring that by distributing the pillars as

far as possible from each other within a roof, as number of

centroids among the gravity weight of data distribution in the

vector space the pillars can withstand the roof’s pressure and stabilize a house or building. Therefore, this algorithm

designates positions of initial centroids in the farthest

accumulated distance between them in the data distribution.

P-ISSN: 2347-4408

E-ISSN: 2347-4734


GLCM Features

Gray co matrix calculates the GLCM from a scaled version of

the image. By default, if I is a binary image, gray co

matrix scales the image to two gray-levels. If I is an intensity

image, gray co matrix scales the image to eight gray-levels.

You can specify the number of gray-levels gray co

matrix uses to scale the image by using the 'Num

Levels' parameter, and the way that gray co matrix scales the

values using the 'Gray Limits' parameter.

The following figure shows how gray co matrix calculates

several values in the GLCM of the 4-by-5 image I. Element

(1,1) in the GLCM contains the value 1 because there is only

one instance in the image where two, horizontally adjacent

pixels have the values 1 and 1. Element (1,2) in the GLCM

contains the value 2 because there are two instances in the

image where two, horizontally adjacent pixels have the

values 1 and2. Gray co matrix continues this processing to fill

in all the values in the GLCM.

For quantitatively describing the first order statistical features

of the image, useful features of the image can be obtained

from the histogram. Mean is the average value of the intensity

of the image. Variance tells the intensity variation around the

means skewness is the measure which tells the symmetries of

the histogram around the mean. Kurtosis is the flatness of the

histogram. Uniformity of the histogram is represented by the

entropy. Following is the list of features obtained using

histogram of the image.

Histogram based features are local in nature. These features

do not consider spatial information into consideration. So for

this purpose gray-level spatial co–occurrence matrix h d(I ,j) based features are defined which are known as second order

histogram based features. These features are based on the

joint probability distribution of pairs of pixels. Distance and

angle _ within a given neighborhood are used for calculation

of joint probability distribution between pixels [21]. Normally

d = 1,2 and 𝜃 = °, °, 9 ° 𝑎𝑛𝑑 °are usedfor calculation. Texture features can be described using this co-occurrence

matrix. Following equations define these features.

P-ISSN: 2347-4408

E-ISSN: 2347-4734


Self-organizing map (SOM) based initial training

A self-organizing map (SOM) or self-organizing

feature map (SOFM) is a type of artificial neural

network (ANN) that is trained using unsupervised learning to

produce a low-dimensional (typically two-dimensional),

discretized representation of the input space of the training

samples, called a map. Self-organizing maps differ from other

artificial neural networks as they apply competitive

learning as opposed to error-correction learning (such as back

propagation with gradient descent), and in the sense that they

use a neighborhood function to preserve

the topological properties of the input space.

A self-organizing map showing U.S. Congress

voting patterns visualized in Synapse. The first two boxes

show clustering and distances while the remaining ones show

the component planes. Red means a yes vote while blue

means a no vote in the component planes (except the party

component where red is Republican and blue is Democratic).

This makes SOMs useful for visualizing low-

dimensional views of high-dimensional data, akin

to multidimensional scaling. The artificial neural network

introduced by the Finnish professor Teuvo Kohonen in the

1980s is sometimes called a Kohonen map or network. The

Kohonen net is a computationally convenient abstraction

building on work on biologically neural models from the

1970s and morphogenesis models dating back to Alan

Turing in the 1950s.

Like most artificial neural networks, SOMs operate

in two modes: training and mapping. "Training" builds the

map using input examples (a competitive process, also called

vector quantization), while "mapping" automatically classifies

a new input vector.

A self-organizing map consists of components called

nodes or neurons. Associated with each node is a weight

vector of the same dimension as the input data vectors, and a

position in the map space. The usual arrangement of nodes is

a two-dimensional regular spacing in

a hexagonal or rectangular grid. The self-organizing map

describes a mapping from a higher-dimensional input space to

a lower-dimensional map space. The procedure for placing a

vector from data space onto the map is to find the node with

the closest (smallest distance metric) weight vector to the data

space vector.

While it is typical to consider this type of network

structure as related to feed forward networks where the nodes

are visualized as being attached, this type of architecture is

fundamentally different in arrangement and motivation.

Useful extensions include using toroidal grids where opposite

edges are connected and using large numbers of nodes.

It has been shown that while self-organizing maps

with a small number of nodes behave in a way that is similar

https://en.wikipedia.org/wiki/Artificial_neural_networkhttps://en.wikipedia.org/wiki/Artificial_neural_networkhttps://en.wikipedia.org/wiki/Unsupervised_learninghttps://en.wikipedia.org/wiki/Competitive_learninghttps://en.wikipedia.org/wiki/Competitive_learninghttps://en.wikipedia.org/wiki/Backpropagationhttps://en.wikipedia.org/wiki/Backpropagationhttps://en.wikipedia.org/wiki/Gradient_descenthttps://en.wikipedia.org/wiki/Topologyhttps://en.wikipedia.org/wiki/Scientific_visualizationhttps://en.wikipedia.org/wiki/Multidimensional_scalinghttps://en.wikipedia.org/wiki/Finlandhttps://en.wikipedia.org/wiki/Teuvo_Kohonenhttps://en.wikipedia.org/wiki/Morphogenesishttps://en.wikipedia.org/wiki/Alan_Turinghttps://en.wikipedia.org/wiki/Alan_Turinghttps://en.wikipedia.org/wiki/Competitive_learninghttps://en.wikipedia.org/wiki/Vector_quantizationhttps://en.wikipedia.org/wiki/Hexagonalhttps://en.wikipedia.org/wiki/Rectangularhttps://en.wikipedia.org/wiki/Feedforward_neural_networkshttps://en.wikipedia.org/wiki/Torushttps://en.wikipedia.org/wiki/File:Synapse_Self-Organizing_Map.png

P-ISSN: 2347-4408

E-ISSN: 2347-4734


to K-means, larger self-organizing maps rearrange data in a

way that is fundamentally topological in character.

It is also common to use the U-Matrix.[5]

The U-

Matrix value of a particular node is the average distance

between the node's weight vector and that of its closest

neighbors. In a square grid, for instance, we might consider

the closest 4 or 8 nodes (the Von Neumann and Moore

neighborhoods, respectively), or six nodes in a hexagonal

grid.

Training is an iterative process which requires a lot

of computational effort and thus is time-consuming. The

extracted features are trained by drawing sample vectors from

the input dataset and ‘teaching’ them to the classifier. The teaching is the process of choosing a winner unit by means of

a similarity measure and updating the values of codebook

vectors in the neighborhood of the winner unit and it is

repeated a number of times. The SOM is a neural network

model belongs to the category of competitive learning

networks and it is predicated on unsupervised learning, that

denotes no human intervention is needed during the learning

but little needs to be known about the characteristics of the

input data. The SOM can be used for clustering data without

knowing the class memberships of the input data and

acclimated to detect features innate to the quandary and thus

has additionally been called self-organizing feature map.

Figure: SOM architecture

In the initial training, the extracted feature vector is drawn

randomly from the input data set, then it is fed to all units in

the network and a similarity measure is calculated between

the input data sample and all the codebook vectors. The

codebook vector is chosen as the best-matching unit (BMU)

with greatest similarity with input sample. The similarity is

usually defined by means of a distance measure, for example

in the case of Euclidean distance the BMU is the closest

neuron to the sample in the input space. Consider ΔZ = {d(z), n = 1, 2, … N} is the extracted MRI feature vectors and the Euclidean norm of the vector d(z) is defined as

Then, the Euclidean distance can define in terms of the

Euclidean norm:

The BMU is the codebook vector usually noted as mc that is a

bestmatch of a given input vector d(z).

After finding the BMU, units in the SOM are updated and the

BMU updated to be a little closer to the sample vector in the

input space. Similarly the topological neighbors of the BMU

are also updated. This update procedure stretches the BMU

and its topological neighbors towards the sample vector. The

update procedure is illustrated in Fig. 3.4 in that the codebook

vectors are situated in the crossings of the solid lines. The

lines drawn in figure represent the topological relationships of

the SOM. The input fed to the network is marked by d(z) in

the input space and the BMU, or the winner neuron is the

codebook vector closest to the sample in the middle above

d(z). The winner neuron and its topological neighbors are

updated by moving them a little towards the input sample,

which is represented in figure by dashed lines. The

neighborhood in this case consists of eight neighboring units

mentioned in the figure.

Figure: updating the best matching unit and its neighbors

The BMU among all the neurons and updating the codebook

vectors in the neighborhood of the winner unit are the major

computational effort of SOM. If the neighborhood is large in

https://en.wikipedia.org/wiki/K-means_algorithmhttps://en.wikipedia.org/wiki/U-Matrixhttps://en.wikipedia.org/wiki/Self-organizing_map#cite_note-UltschSiemon1990-5https://en.wikipedia.org/wiki/Von_Neumann_neighborhoodhttps://en.wikipedia.org/wiki/Moore_neighborhoodhttps://en.wikipedia.org/wiki/Moore_neighborhood

P-ISSN: 2347-4408

E-ISSN: 2347-4734


the beginning of the training process, then lot of codebook

vectors to be updated and if the network is large, then

relatively larger portion of the time is spent for looking a

winner neuron. Time to be spent on each of these phases

depends on particular software and hardware used.

Classification:

In machine learning and statistics, classification is the

problem of identifying to which of a set of categories (sub-

populations) a new observation belongs, on the basis of

a training set of data containing observations (or instances)

whose category membership is known.

An example would be assigning a given email into "spam" or

"non-spam" classes or assigning a diagnosis to a given patient

as described by observed characteristics of the patient

(gender, blood pressure, presence or absence of certain

symptoms, etc.). Classification is an example of pattern

recognition.

In the terminology of machine learning,

classification is

considered an instance of supervised learning, i.e. learning

where a training set of correctly identified observations is

available. The corresponding unsupervised procedure is

known as clustering, and involves grouping data into

categories based on some measure of inherent similarity

or distance.

Figure : Concept of machine learning based classification

Often, the individual observations are analyzed into a set of

quantifiable properties, known variously as explanatory

variables or features. These properties may variously

be categorical (e.g. "A", "B", "AB" or "O", for blood

type), ordinal (e.g. "large", "medium" or "small"), integer-

valued (e.g. the number of occurrences of a particular word in

an email) or real-valued (e.g. a measurement of blood

pressure). Other classifiers work by comparing observations

to previous observations by means of a similarity or distance

function.

An algorithm that implements classification, especially in a

concrete implementation, is known as a classifier. The term

"classifier" sometimes also refers to the

mathematical function, implemented by a classification

algorithm, that maps input data to a category.

Terminology across fields is quite varied. In statistics, where

classification is often done with logistic regression or a

similar procedure, the properties of observations are

termed explanatory variables (or independent variables,

regressors, etc.), and the categories to be predicted are known

as outcomes, which are considered to be possible values of

the dependent variable. In machine learning, the observations

are often known as instances, the explanatory variables are

termed features (grouped into a feature vector), and the

possible categories to be predicted are classes. Other fields

may use different terminology: e.g. in community ecology,

the term "classification" normally refers to cluster analysis.

Support Vector Machine Classifier

Support vector machines (SVMs) are a set of supervised

learning methods used for classification, regression and

outliers detection.More formally, a support vector machine

constructs a hyperplane or set of hyper planes in a high- or

infinite-dimensional space, which can be used for

classification, regression, or other tasks. Intuitively, a good

separation is achieved by the hyperplane that has the largest

distance to the nearest training-data point of any class (so-

called functional margin), since in general the larger the

margin the lower the generalization error of the classifier.

SVM Algorithm

Classifying data is a common task in machine learning.

Suppose some given data points each belong to one of two

classes, and the goal is to decide which class a new data point

will be in. In the case of support vector machines, a data point

is viewed as a p-dimensional vector , and we want to know

whether we can separate such points with a (p-1)-dimensional

hyperplane. This is called a linear classifier. There are many

hyperplanes that might classify the data. One reasonable

choice as the best hyperplane is the one that represents the

largest separation, or margin, between the two classes. So we

https://en.wikipedia.org/wiki/Machine_learninghttps://en.wikipedia.org/wiki/Statisticshttps://en.wikipedia.org/wiki/Categorical_datahttps://en.wikipedia.org/wiki/Observationhttps://en.wikipedia.org/wiki/Training_sethttps://en.wikipedia.org/wiki/Spam_filteringhttps://en.wikipedia.org/wiki/Spam_filteringhttps://en.wikipedia.org/wiki/Pattern_recognitionhttps://en.wikipedia.org/wiki/Pattern_recognitionhttps://en.wikipedia.org/wiki/Supervised_learninghttps://en.wikipedia.org/wiki/Unsupervised_learninghttps://en.wikipedia.org/wiki/Cluster_analysishttps://en.wikipedia.org/wiki/Distancehttps://en.wikipedia.org/wiki/Explanatory_variableshttps://en.wikipedia.org/wiki/Explanatory_variableshttps://en.wikipedia.org/wiki/Categorical_datahttps://en.wikipedia.org/wiki/Blood_typehttps://en.wikipedia.org/wiki/Blood_typehttps://en.wikipedia.org/wiki/Ordinal_datahttps://en.wikipedia.org/wiki/Integerhttps://en.wikipedia.org/wiki/Integerhttps://en.wikipedia.org/wiki/Emailhttps://en.wikipedia.org/wiki/Real_numberhttps://en.wikipedia.org/wiki/Blood_pressurehttps://en.wikipedia.org/wiki/Blood_pressurehttps://en.wikipedia.org/wiki/Similarity_functionhttps://en.wikipedia.org/wiki/Algorithmhttps://en.wikipedia.org/wiki/Function_(mathematics)https://en.wikipedia.org/wiki/Statisticshttps://en.wikipedia.org/wiki/Logistic_regressionhttps://en.wikipedia.org/wiki/Explanatory_variablehttps://en.wikipedia.org/wiki/Independent_variablehttps://en.wikipedia.org/wiki/Dependent_variablehttps://en.wikipedia.org/wiki/Feature_vectorhttps://en.wikipedia.org/wiki/Community_ecologyhttps://en.wikipedia.org/wiki/Cluster_analysis

P-ISSN: 2347-4408

E-ISSN: 2347-4734


choose the hyper plane so that the distance from it to the

nearest data point on each side is maximized. If such a

hyperplane exists, it is known as the maximum-margin

hyperplane and the linear classifier it defines is known as a

maximum margin classifier; or equivalently, the perception of

optimal stability.

Figure : Linearly Separable patterns

Applications

SVMs are helpful in text and hypertext categorization as their

application can significantly reduce the need for labeled

training instances in both the standard inductive and

transductive settings.

Classification of images can also be performed using

SVMs. Experimental results show that SVMs achieve

significantly higher search accuracy than traditional query

refinement schemes after just three to four rounds of

relevance feedback. This is also true of image segmentation

systems, including those using a modified version SVM that

uses the privileged approach as suggested by Vapnik. The

SVM algorithm has been widely applied in the biological and

other sciences. They have been used to classify proteins with

up to 90% of the compounds classified correctly. Permutation

tests based on SVM weights have been suggested as a

mechanism for interpretation of SVM models. Support vector

machine weights have also been used to interpret SVM

models in the past. Posthoc interpretation of support vector

machine models in order to identify features used by the

model to make predictions is a relatively new area of research

with special significance in the biological sciences.

Advantages of support vector machines are:

Effective in high dimensional spaces. Still effective in cases where number of dimensions

is greater than the number of samples.

Uses a subset of training points in the decision function (called support vectors), so it is also

memory efficient.

Versatile: different Kernel functions can be specified for the decision function. Common kernels are

provided, but it is also possible to specify custom

kernels.

Disadvantages of support vector machines:

If the number of features is much greater than the number of

samples, the method is likely to give poor performances.

SVMs do not directly provide probability estimates, these are

calculated using an expensive five-fold cross-validation.

5. SCREENSHOTS

The proposed system is implemented using a

MATLAB 2014a.The algorithm is tested on a BRAINX

database of 397 images where it is evaluated for brain tumor

detection. These images are the most widely used standard

test images used for image retargeting algorithms. The images

contains a nice mixture of detail, flat regions, shading and

texture that do a good job of testing various image processing

algorithms. These images are used for many image processing

researches.

Figure :Test images (Top row: normal images & Bottom

row: Abnormal images)

P-ISSN: 2347-4408

E-ISSN: 2347-4734


This section elaborates the results obtained for the proposed

work and the proposed work is implemented on MATLAB

GUI shown in below figures.

Figure :GUI for Pre-processing

Figure : GUI with Layer segmentation

The following table 4.1 presenting the features distribution of

normal and tumorous images in this proposed work.

Table : Feature matrix of 30 sample

Figure : GUI for Two-Tier classification

The accuracy of the proposed work is show in the following

plots.

6. CONCLUSION

This proposed work genesis an efficient recognition

system for the brain tumor classification and the by not

focusing on the traditional way, this work travels on two tier

classification method. This work presented for totally 60

images of both normal and abnormal images and MATLAB

image processing toolbox is useful for developing the

proposed work. Firstly the brain MRI images are pre-

processed because it is well-known to every one about the

noises present in the MRI images. The pre-processed images

are undergone for the segmentation where the brain ROI

extracted precisely. The feature extraction is done by

processing tithe GLC matrix and the shape and texture

features are collected for every image. The collected features

vectors are trained with the help of SOM neural network

where the variance among the feature set is increased tightly

and making the classifier learning rate as high. The SVM

classifier is effectively classifying the images very accurately

and the also it assures maximum accuracy as 89.5%.

7. FUTURE WORK

This proposed work is focusing the traditional parametric

feature extraction technique for this recognition task. The

future work may be extended this feature extraction process in

terms of deep learning neural network. Convolution neural

networks are the most delicate network for the field of image

processing and also the algorithm need to be extended for

segmenting the tumorous area.

P-ISSN: 2347-4408

E-ISSN: 2347-4734


REFERENCE

[1] U. Vovk, F. Pernus, and B. Likar, “A review of methods for correction of intensity in homogeneity in MRI,” IEEE Trans .Med. Imag. vol. 26, no. 3, pp. 405–421, Mar. 2007. [2] W. Wells, E. Grimson, R. Kikinis, and F. Jolesz, “Adaptive segmentation of MRI data,” IEEE Trans. Med. Imag., vol. 15, no. 4, pp. 429–442, Apr. 1996. [3] V. Leemput, K. Maes, D. Vandermeulen, and P. Suetens,

“Automated model-based bias field correction of MR images of the brain,” IEEE Trans. Med. Imag., vol. 18, no. 10, pp. 885–896, Oct. 1999.

[4] Y. Zhang, M. Brady, and S. Smith, “Segmentation of brain MR images through a hidden Markov random field model and the

expectation maximization algorithm,” IEEE Trans .Med. Imag., vol. 20, no. 1, pp. 45– 57, Jan. 2001. [5] C. Li, C. Gatenby, L. Wang, and J. Gore, “A robust parametric method for bias field estimation and segmentation of MR images,” in Proc. IEEE Conf. Comput. Vision Pattern Recog., 2009, pp. 218–223.

[6] C. Li, C. Xu, A. Anderson, and J. Gore, “MRI tissue classification and bias field estimation based on coherent local

intensity clustering: A unified energy minimization framework,” in Proc. 21st Int. Conf. Inf. Process. Med. Imag., Lecture Notes in

Computer Science, 2009, vol. 5636, pp. 288– 299. [7] K. Sikka, N. Sinha, P. K. Singh, and A. K. Mishra, “A fully automated algorithm under modified FCM framework for improved

brain MR image segmentation,” Magn. Reson. Imag., vol. 27, pp. 994–1004, Jul. 2009. [8] D. Pham and J. Prince, “Adaptive fuzzy segmentation of magnetic resonance images,” IEEE Trans.Med. Imag., vol. 18, no. 9, pp. 737–752, Sep. 1999. [9] M. Ahmed, S. Yamany, N. Mohamed, A. Farag , and T. Moriarty, “A modified fuzzy c-means algorithm for bias field estimation and segmentation of MRI data,” IEEE Trans. Med. Imag., vol. 21, no. 3, pp. 193–199, Mar. 2002. [10] A. Liew and H. Yan, “An adaptive spatial fuzzy clustering algorithm for 3-D MR image segmentation,” IEEE Trans. Med. Imag., vol. 22, no. 9, pp. 1063–1075, Sep. 2003. [11] Z. X. Ji, Q. Chen, Q. S. Sun, D. S. Xia, and P. A. Heng, “MR image segmentation and bias field estimation using coherent local

and global intensity clustering,” in Proc. 7th Int. Conf. Fuzzy Syst. Knowl. Discov., 2010, vol. 2, pp. 578–582. [12] Z. X. Ji, Q. S. Sun, and D. S. Xia, “A modified possibilistic fuzzy c-means clustering

algorithm for bias field estimation and segmentation of brain MR

image,” Comput. Med. Imag. Graph., vol. 35, pp. 383–397, Jul. 2011.

[13] S. C. Chen and D. Q. Zhang, “Robust image segmentation using FCM with spatial constraints based on new kernel-induced distance

measure,” IEEE Trans. Syst. Man Cybernet., vol. 34, no. 4, pp. 1907–1916, Apr. 2004.

[14] M. S. Yang and H. S. Tsai, “A Gaussian kernel-based fuzzy c-means algorithm with a spatial bias correction,” Pattern Recog. Lett., vol. 29, pp. 1713–1725, Sep. 2008. [15] L. Liao, T. Lin, and B. Li, “MRI brain image segmentation and bias field correction based on fast spatially constrained kernel

clustering approach,” Pattern Recog. Lett., vol. 29, pp. 1580–1588, Jul. 2008.

[16] B. Vijayakumar, Ashish Chaturvedi , Brain Tumor in Three

Dimensional Magnetic Resonance Images and Concavity Analysis,

International Journal of Computer Application, Issue 3, Volume 1

(February 2013),

[17] L. Gupta and T. Sortrakul, “A Gaussian mixture based image segmentation algorithm,” Pattern Recog., vol. 31, pp. 315–325, Mar. 1998.

[18] H. Greenspan, A. Ruf, and J. Goldberger, “Constrained Gaussian mixture model framework for automatic segmentation of

MRI brain images,” IEEE Trans. Med. Imag., vol. 25, no. 9, pp. 1233–1245, Sep. 2006. [19] K. Blekas, A. Likas, N. P. Galatsanos, and I. E. Lagaris, “A spatially constrained mixture model for image segmentation,” IEEE Trans. Neural Netw., vol. 16, no. 2, pp. 494–498, Feb. 2005. [20] D. Tran, T. Le, and M. Wagner, “Fuzzy Gaussian mixture models for speaker recognition,” in Proc. Int. Conf. Spoken Language, 1998, pp. 759– 762. [21] J. Zeng, L. Xie, and Z. Liu, “Type-2 fuzzy Gaussian mixture,” Pattern Recog., vol. 41, pp. 3636–3643, Dec. 2008. [22] J. C. Bezdek, Pattern Recognition With Fuzzy Objective

Function Algorithms. Norwell, MA: Kluwer, 1981.

[23] L. Wang, L. He, A. Mishra, and C. Li, “Active contours driven by local Gaussian distribution fitting energy,” Signal Process., vol. 89, pp. 2435– 2447, Dec. 2009. [24] Y. J. Chen, J. W. Zhang, and J. Macione, “An improved level set method for brain MR images segmentation and bias correction,” Comput. Med. Imag. Graph., vol. 33, pp. 510–519, May 2009. [24] B. Vijayakumar, Ashish Chaturvedi, Effective Classification of

Anaplastic Neoplasm in Huddling Stain Image by Fuzzy Clustering

Method, International Journal of Scientific Research, Issue 3 volume

2, March-April 2013,

[25] Vijayakumar, B., and Ashish Chaturvedi. "Automatic Brain

Tumors Segmentation of MR Images using Fluid Vector Flow and

Support Vector Machine." Research Journal of Information

Technology 4.

[26] Vijayakumar, B., and Ashish Chaturvedi. "Tumor Cut-

Segmentation and Classification of MR Images using Texture

Features and Feed Forward Neural Networks." European Journal of

Scientific Research 85.3 (2012): 363-372

[27] Vijayakumar, B., and Ashish Chaturvedi, “Idiosyncrasy Dissection and Assorting of Brain MR Images Instigating Kernel

Corroborated Support Vector Machine,” Archive Des Science

P-ISSN: 2347-4408

E-ISSN: 2347-4734


[28] Vijayakumar, B., and Ashish Chaturvedi, “Abnormality segmentation and classification of brain MR images using Kernel

based Support vector machine,” Archive Des Science,Volume 66,Issue 4.

[29]. B.Balakumar, P.Raviraj, “An Automatic Brain Tumor Detection and Segmentation Scheme for Clinical Binary Images”, International Journal of Emerging Technologies in Computational

and Applied Sciences(IJETCAS), Vol.8(1), May 2014, pp.37-42

(Impact factor 0.30)

[30]B.Balakumar, P.Raviraj, “Kernel Analysis Mercers Theorem on Matrix Model Ependymomas Tumor in Brain Images”, International Review on Computers and Software,(IROCS), Vol.9, No.6, June

2014, pp.997-1007 (Impact Factor 0.46)

[31] Qurat-Ul-Ain, GhazanfarLatif, et al. "Classification and

segmentation of brain tumor using texture analysis." Recent

Advances In Artificial Intelligence, Knowledge Engineering And

Data Bases (2010): 147-155.

[32] Khalid, Noor Elaiza Abdul, Shafaf Ibrahim, and P. N. M. M.

Haniff. "MRI Brain Abnormalities Segmentation using K-Nearest

Neighbors (k-NN)."International Journal on Computer Science and

Engineering 3.2 (2011): 980-990.

[33] Aslam, Hakeem Aejaz, TirumalaRamashri, and Mohammed

Imtiaz Ali Ahsan. "A new approach to image segmentation for brain

tumor detection using pillar K-means algorithm." International

Journal of Advanced Research in Computer and Communication

Engineering 2.3 (2013).

[34] Anbeek, Petronella, Koen L. Vincken, and Max A. Viergever.

"Automated MS-lesion segmentation by k-nearest neighbor

classification." MIDAS J(2008).

[35] Padma, A., and Dr R. Sukanesh. "Automatic diagnosis of

abnormal tumor region from brain computed tomography images

using wavelet based statistical texture features." arXiv preprint

arXiv: 1109.1067 (2011).

[36] Steenwijk, Martijn D., et al. "Accurate white matter lesion

segmentation by k nearest neighbor classification with tissue type

priors (kNN-TTPs)."NeuroImage: Clinical 3 (2013): 462-469.

[37] Barakbah, Ali Ridho, and Yasushi Kiyoki. "A pillar algorithm

for k-means optimization by distance maximization for initial

centroid designation. "Computational Intelligence and Data Mining,

2009. CIDM'09. IEEE Symposium on. IEEE, 2009.

[38] Xu, Yong, and Fengxi Song. "Feature extraction based on a

linear separability criterion." International Journal of Innovative

Computing, Information and Control 4.4 (2008): 857-865.

[39] Ain, Quratul, M. Arfan Jaffar, and Tae-Sun Choi. "Fuzzy

anisotropic diffusion based segmentation and texture based ensemble

classification of brain tumor." Applied Soft Computing 21 (2014):

330-340.

[40] Ohno, Asako, and Hajime Murao. "A similarity measuring

method for images based on the feature extraction algorithm using

reference vectors."Innovative Computing, Information and Control,

2007. ICICIC'07. Second International Conference on. IEEE, 2007.

[41] Dehghani, Sara, and MashallahAbbasiDezfooli. "Breast cancer

diagnosis system based on contour let analysis and support vector

machine." World Applied Sciences Journal 13.5 (2011): 1067-1076.

[42] Grau, Vicente, et al. "Improved watershed transform for medical

image segmentation using prior information." IEEE transactions on

medical imaging23.4 (2004): 447-458.

[43] Jaffar, M. Arfan, et al. "Fuzzy entropy and morphology based

fully automated segmentation of lungs from CT scan images."

International Journal of Innovative Computing, Information and

Control 5.12 (2009): 4993-5002.

[44] Zhang, Yongyue, Michael Brady, and Stephen Smith.

"Segmentation of brain MR images through a hidden Markov

random field model and the expectation-maximization algorithm."

IEEE transactions on medical imaging20.1 (2001): 45-57.

[45] Li, Bin, Wufan Chen, and Dandan Wang. "An improved FCM

algorithm incorporating spatial information for image segmentation."

Computer Science and Computational Technology, 2008.

ISCSCT'08. International Symposium on. Vol. 2. IEEE, 2008.

brain tumor classification using machine learning algorithms. brain-tumor... · 2018. 7. 31. ·...

Documents