UNSUPERVISED LEARNING TECHNIQUES WITH MRI BRAIN IMAGE USING GUI MATLAB TOOLBOX

(1)

ISSN: 2005-4238 IJAST 471

UNSUPERVISED LEARNING TECHNIQUES WITH MRI BRAIN IMAGE USING GUI MATLAB TOOLBOX

Sudha Tiwari¹, Dr. Samarendra Mohan Ghosh²,Aradhana³

1,3Research Scholar in Dr. C.V. Raman University Bilaspur,

2Professor in Dr. C.V. Raman University Bilaspur

Abstract

In many application areas Image processing is used. The objective of image segmentation is to identify the feature of image. In this work it is divided into some part preprocessing, segmentation, execution comparison and expected result. Soft computing is an intelligence approach found that result it works automatically give result simultaneously at a time. Soft Computing gives better response and MATLAB gives better results than manual process. Automatic process is done by Soft Computing approach against manual process. Needs to improve the accuracy of tumor detection.

Keyword: MRI Image, segmentation, Soft Computing and MATLAB GUI TOOLBOX.

1. Introduction

Identification and detection of tumor is done by physical expert. Brain tumor is defined as the abnormal growth of cells. Brain tumor is an abnormal mass of tissue. To extract useful information from images with the help of computer system by MRI.

2. MRI

Soft tissue of human body is identified by MRI scene. It is really a technique that works on the magnetic field and radio waves to create detailed images of the organs and tissues within your body. MRI is widely used to visualize brain structures such as for instance white matter, grey matter, and ventricles and to detect abnormalities. The MRI may be the usually used modality for brain tumor growth imaging and location finding. It is really a medical imaging technique used to imagine the internal structure of the human body and offer high quality images. MRI supplies a greater distinctive between different tissues of the body. MRI contains useful and good information that may be used in improving the grade of diagnosis and treatment of brain. MRI image texture holds rich sources of information such as for instance characterize brightness, color, slope, size, and other features. Most MRI machines are large, tube-shaped magnets.

3. Methodology

This process is based on unsupervised learning technique.

3.1 Preprocessing-

(1) Input image –To find better solution of this approach we need to iterate the process. The input brain image is first required to identify the tumor area

(2) Gray scale image (3) Binary image

(4) Denoising- there are defferent types of filtering techniques is available. In this work wiener filter is used to remove the noise.

3.2 Segmentation

Segmentation is a powerful tool to detect the tumor area in brain. Image segmentation is the process of partitioning a digital image into multiple segments which consist of sets of pixels. Image segmentation is basically used to locate objects and boundaries (lines, curves, etc.) in a particular image. More precisely, image segmentation is the process of assigning a label to each and every pixel in an image such that the pixels with the same label share certain visual characteristics. The goal of segmentation is to simplify and/or change the representation of an image into something that is easier and more meaningful to analyze. In case of medical image segmentation, the aim is to:

• Study anatomical structure.

(2)

ISSN: 2005-4238 IJAST 472

• Identify the Region of Interest,i.e., locate tumor,lesion and any other abnormalities.

• Measure tissue volume to measure growth of tumor (also decrease in size of tumor with treatment).

3.3 Types of segmentation

(1) Edge method-Method of edge detection thechnique- sobel, Canny, Prewitt, Robert and LoG edge detection.

(2) Threshold method-Thresholding is a property of segmentation based on intensity value of pixel.

(3) Region method- It is the set of connected pixel called region.

(4) Clustering method-

(a) Fuzzy C-means- It is unsupervised learning technique in which dataset is grouped into n cluster with every data point in data set that belongs to a every cluster to a certain degree.

This algorithm works by assigning membership to each data point corresponding to each cluster center on the basis of distance between the cluster center and the data point. More the data is near to the cluster center more is its membership towards the particular cluster center. Clearly, summation of membership of each data point should be equal to one. After each iteration membership and cluster centers are updated according to the formula:

where,

'n' is the number of data points. 'vj' represents the j^th cluster center. 'm' is the fuzziness index m € [1, ∞]. 'c' represents the number of cluster center. 'µij' represents the membership of i^th data to j^th cluster center. 'dij' represents the Euclidean distance between i^th data and j^th cluster center.

Main objective of fuzzy c-means algorithm is to minimize:

where,

'||xi – vj||' is the Euclidean distance between i^th data and j^th cluster center.

Algorithmic steps for Fuzzy c-means clustering

Let X = {x1, x2, x3 ..., xn} be the set of data points and V = {v1, v2, v3 ..., vc} be the set of centers.

(1) Randomly select ‘c’ cluster centers.

(2) Calculate the fuzzy membership 'µij' using:

(3) Compute the fuzzy centers 'vj' using:

Advantages

1) Gives best result for overlapped data set and comparatively better then k-means algorithm.

2) Unlike k-means where data point must exclusively belong to one cluster center here data point is assigned membership to each cluster center as a result of which data point may belong to more then one

cluster center.

Disadvantages

(1) specification of the number of clusters.

(2) With lower value of β we get the better result but at the expense of more number of iteration.

(3) Euclidean distance measures can unequally weight underlying factors.

(4) Repeat step 2) and 3) until the minimum 'J' value is achieved or ||U^(k+1)- U^(k)|| < β. where, ‘k’ is the iteration step. B ‘β’ is the termination criterion between [0, 1].‘U = (µij)n*c’ is the fuzzy membership matrix. J’ is the objective function

(3)

ISSN: 2005-4238 IJAST 473

(b) Fuzzy K-means- k-means is one of the simplest unsupervised learning algorithms that solve the well known clustering problem. The procedure follows a simple and easy way to classify a given data set through a certain number of clusters (assume k clusters) fixed apriori. The main idea is to define k centers, one for each cluster. These centers should be placed in a cunning way because of different location causes different result. So, the better choice is to place them as much as possible far away from each other. The next step is to take each point belonging to a given data set and associate it to the nearest center. When no point is pending, the first step is completed and an early group age is done.

At this point we need to re-calculate k new centroids as barycenter of the clusters resulting from the previous step. After we have these k new centroids, a new binding has to be done between the same data set points and the nearest new center. A loop has been generated. As a result of this loop we may notice that the k centers change their location step by step until no more changes are done or in other words centers do not move any more.

objective function know as squared error function that is

where,

‘||xi - vj||’ is the Euclidean distance between xi and vj.

‘ci’ is the number of data points in i^th cluster.

‘c’ is the number of cluster centers.

Algorithmic steps for k-means clustering

Let X = {x1,x2,x3,……..,xn} be the set of data points and V = {v1,v2,…….,vc} be the set of centers.

1) Randomly select ‘c’ cluster centers.

2) Calculate the distance between each data point and cluster centers.

3) Assign the data point to the cluster center whose distance from the cluster center is minimum of all the cluster centers..

4) Recalculate the new cluster center using:

where, ‘ci’ represents the number of data points in i^th cluster.

5) Recalculate the distance between each data point and new obtained cluster centers.

6) If no data point was reassigned then stop, otherwise repeat from step 3).

Advantages

(1) Fast, robust and easier to understand.

(2) Relatively efficient: O(tknd), where n is # objects, k is # clusters, d is # dimension of each object, and t is # iterations. Normally, k, t, d << n.

(3) Gives best result when data set are distinct or well separated from each other.

Fig 1: Showing the result of k-means for 'N' = 60 and 'c' = 3.

Disadvantages

(1) The learning algorithm requires apriori specification of the number of cluster centers.

(2) The use of Exclusive Assignment - If there are two highly overlapping data then k-means will not be able to resolve that there are two clusters.

(4)

ISSN: 2005-4238 IJAST 474

(3) The learning algorithm is not invariant to non-linear transformations i.e. with different representation of data we get different results (data represented in form of cartesian co-ordinates and polar co-ordinates will give different results).

(4) Euclidean distance measures can unequally weight underlying factors.

5) The learning algorithm provides the local optima of the squared error function.

6) Randomly choosing of the cluster center cannot lead us to the fruitful result. Pl. refer Fig.

7) Applicable only when mean is defined i.e. fails for categorical data.

8) Unable to handle noisy data and outliers.

9) Algorithm fails for non-linear data set.

3.4 Soft Computing:

Soft Computing is different from hard computing. It is tolerant of the following.

-Imprecision -Uncertainty -Partial truth -Approximation.

It uses these all as input. The role model of soft computing is human mind. The guidance principal of soft computing is Exploit the tolerance for Imprecision, Uncertainty, Partial truth, Approximation to achieve robustness and low cost solution. SC is emerging field of Computer Science where no sufficient information is available for exact solution for given problem. It is useful for decision making task and deals with assumption. It posses the simulation of human behavior and cognitive process on a computer.

3.5 The principal component includes:

-Genetic algorithms -Artificial Intelligence -Fuzzy logic

-Neural Network

It refers to complex problem to solution that is unpredictable, uncertain and between 0 and 1. Soft Computing is an emerging field that consists of complementary elements of fuzzy logic, neural computing and evolutionary computation. One of the most important applications is image segmentation in field of The process of partitioning a digital image into multiple regions or sets of pixels is called image segmentation. Segmentation is an essential step in image processing since it conditions the quality of the resulting interpretation. Lots of approaches have been proposed and a dense literature is available In order to extract as much information as possible from an environment, multicomponent images can be used. In the last decade, multicomponent images segmentation has received a great deal of attention for soft computing applications because it significantly improves the discrimination and the recognition capabilities compared with gray-level image segmentation methods. In this paper, the main aim is to understand the soft computing approach to image segmentation using MATLAB TOOLBOX.

4. GUI MATLAB TOOLBOX

This toolbox provides tools to create sophisticated MATLAB graphical user interfaces that resize

gracefully. The classes supplied can be used in combination to produce virtually any user interface layout.

* Arrange MATLAB user-interface components horizontally, vertically or in grids

* Ability to mix fixed size and variable size elements

* Dynamic resizing of elements by dragging dividers

* Use panels and tabs for switching interface pages

Documentation is included and is integrated into the MATLAB Help Browser. It can also be viewed as a st For added control over design and development, you can also create MATLAB code that defines all component properties and behaviors. MATLAB contains built-in functionality to help you create the GUI for your app programmatically. You can add dialog boxes, user interface controls (such as push buttons and sliders), and containers (such as panels and button groups).and-alone web-page.

(5)

ISSN: 2005-4238 IJAST 475

GUI programmatically, on the other way, gives all control of design style and features. You can add whatever text windows, push buttons, panels and plots. Those graphic objects will be generated once you run your code. You can also unleash the more advanced features like ‘uitree’ function group which is not available from GUIDE.

It’s nevertheless without pains. You need to build a UI from scratch. Position and size of each UI element have to be specified. Calculations of positions can be extremely painful especially at the early stage of design and constant revisions are needed. Fortunately, with the powerful GUI Layout Toolbox, we don’t have to do all these dirty work by ourselves.

4.1 GUI Layout Toolbox

Developed by David Sampson and Ben Tordoff from the Consulting Services group at MathWorks, this toolbox provides tools to create sophisticated MATLAB graphical user interfaces that resize gracefully. It fits perfectly with built-in MATLAB UI component. In addition, it allows drag-and-drop style of component rearrangement and comes with a highly-requested tab component.

It comes with a well-written concise documentation loaded with examples. The Getting Started section gives you a quick idea of using this toolbox.

4.2 Compare

Fuzzy C-Means and K-Means is compared for their computing performance.

Some researcher found that K-Means is fast, robust and easy to implement. Fuzzy C-Means is a advance and an extension of K-Means. In this paper we compared the efficiency of K-Means and C-Means and we found that all the time C-Means gives better result. Difference between fcm and kcm is, In K-means k centres are defined, one for each cluster. k-means- It is used to divide a group of data points into clusters where in points inside one cluster are similar to each other.Faster, because order of time complexity is linear with the number of data. It is simple to implement, Scales to large data sets. Can implement this of different shapes and size. It is one of the most robust method.

5. Conclusion

There are no of research papers available for each step. In this paper proposed the method of clustering techniques using MATLAB GUI TOOLBOX. GUI gives better feature of images.

References

1. Abdullah, H. N and. Habtr, M. A. (2015), Brain Tumor Extraction Approach in MRI Images Based on Soft Computing Techniques, 8th International Conference on Intelligent Networks and Intelligent Systems, IEEE, November.

2. Ali, S. M., Abood, L. K. and Abdoon, R. S. (2013), Brain Tumor Extraction in MRI images using Clustering and Morphological Operations Techniques, International Journal of Geographical Information System Application And Remote Sensing, ISSN. No. 2077-771X, Vol.-4, No.-1, June.

3. Azhari, E. M., Mudzakkir, M., Hatta, M., Htike, Z. Z. and Lei Win, S. (2014), Brain Tumor Detection And Localization In Magnetic Resonance Imaging, International Journal of Information Technology Convergence and Services (IJITCS), Vol.4, No.1, February, pp. no. 1-11.

4. Bahadure, N. B., Ray, A.K. and Thethi, H.P. (2016), Performance Analysis of Soft Computing and Image Processing Based Image Segmentation”, International Conference on Engineering Technologies and Big Data Analytics (ETBDA), Jan. 21-22, Bangkok (Thailand).

5. Banchpalliwar, R. A. and Dr. Salankar, S. S. (2016), A Review on B rain MRI Image Segmentation Clustering Algorithm, (IOSR-JECE) e-ISSN: 2278-2834, p-ISSN: 2278-8735.Volume 11, Issue 1, Ver. III, Jan. –Feb, PP 80-84.

6. Baraiya, N. and Modi, H. (2016), Comparative Study of Different Methods for Brain Tumor Extraction from MRI Images using Image Processing” , IJST, Vol. 9(4), DOI:

10.17485/ijst/2016/v9i4/85624, January, ISSN (Print) : 0974-6846, ISSN (Online) : 0974-5645.

7. Basak, R., Chakraborty, S., Mondal, A. K. and Biswas, S. B. (2018), Image Segmentation Techniques:

A Survey, International Research Journal of Engineering and Technology, Volume: 05 Issue: 04, Apr., e-ISSN: 2395-0056, p-ISSN: 2395-0072.

(6)

ISSN: 2005-4238 IJAST 476

8. Bora, D. J. and Dr. Gupta, A. K. (2014), A Comparative Study between Fuzzy Clustering Algorithm and Hard Clustering Algorithm, IJCTT, Vol. 10 No. 2 –Apr.

9. Chaudhari, A. K. and Kulkarni, J. V. (2013), Local Entropy Based Brain MR Image Segmentation, 3rd IEEE International Advance Computing Conference (IACC), Pp 1229-1232.

10. Datta, S. and Dr. Chakraborty, M. (2011), Brain Tumor Detection from Pre-Processed MR Images using Segmentation Techniques, IJCA Special Issue on 2^nd National Conference- Computing, Communication and Sensor Network, CCSN, pp. no. 1-5.

11. Devi, N. and Bhattacharyya, A. (2018), A Soft Computing Framework for Brain Tumor Detection through MRI Images, ADBU-Journal of Engineering Technology, AJET, ISSN: 2348-7305, Volume 4(1).

12. Dhopte, R. A. and Ali, Z. (2013), Recent Trends and Applications of Soft Computing: A Survey, IJCSA, Vol. 6, No.2, Apr., ISSN: 0974-1011.

13. Dhanalakshmi, P. and Kanimozhi, T. (2013), Automatic Segmentation of Brain Tumor using K- Means Clustering and its Area Calculation, International Journal of Advanced Electrical and Electronics Engineering (IJAEEE), ISSN (Print): 2278-8948, Volume-2, Issue-2, pp. no. 130-134.

14. Dhurkunde, S. and Patil, S. (2016), Segmentation of Brain Tumor in Magnetic Resonance Images using Various Techniques, IJIRSET. 5, Issue 1, ISSN (Online): 2319-8753, ISSN (Print): 2347-6710, Januray.

15. Ghosh, S. and Dubey, S. K. (2015), Comparative Analysis of K-Means and Fuzzy C-Means Algorithms”, IJACSA, Vol. 4, No.4.

16. Gumaste1, P. P. and Dr. Jadhav, D. V. (2015), Image Segmentation Techniques for Brain MRI Images:-A Survey, IJMTER, e-ISSN No.:2349-9745, Date: 2-4 July.