You are on page 1of 6

AN EFFICIENT APPROACH TO PREDICT TUMOR IN 2D BRAIN IMAGE USING CLASSIFICATION TECHNIQUES

*K.S.Deepak, *K.Gokul, *R.Hinduja,**S.Rajkumar Email id:hinduja24@gmail.com [*Final year students BE-Computer Science And Engineering,K.S.Rangasamy College Of Technology,Truchengode **Assistant Professor K.S.Rangasamy College Of Technology,Truchengode]

ABSTRACT
The Brain is one of the most important organs in human body. The brain controls everything : sight, hearing, taste, touch, emotions etc. In medical field, brain plays an important role in every aspect. In the last decade one of the dangerous diseases is brain tumour and also prediction of tumour in brain is very difficult process. This proposed system explains how to find the brain tumour in patients body using some data mining techniques such as segmentation and classification. Images are considered as one of the most important medium of conveying information. Understanding images and extracting the information from them such that the information can be used for other tasks is an important aspect of Machine learning. One of the first steps in direction of understanding images is to segment them and find out different objects in them. For segmentation we are using K-means clustering algorithm. In the second step we perform classification of MRI brain image using decision tree and SVM classification algorithms and predict which is better classification technique and extract tumour parts in brain. The proposed technique is used to find

diseased brain image of patient to match with the database tumour image.

INTRODUCTION
Image processing is any form of signal processing for which the input is an image and produce the output may be either an image or a set of characteristics or parameters related to the image. Many techniques are adopted for processing, segmenting and retrieving the image. One of the biggest advantages of image processing is the ability of the operator to post-process the image. Post-processing of the image allows the operator to manipulate the pixel shades to correct image density and contrast, as well as perform other processing functions. Data mining (sometimes called data or knowledge discovery) is a process of
searching, cleaning, collecting analyzing data from various sources of databases for the purpose of evaluation. Data mining is the

process of finding correlations or patterns among dozens of fields in large relational databases. For example, this technique will help human experts in quantifying, localizing, and tracking of disease progression among various subjects and within a specific semantic area of the brain.

The proposed system consists of two different processes and also finding the best classification algorithm for image retrieval in medical mining. In first phase, Image segmentation is to partition an image into meaningful regions with respect to a particular application. In Second phase Classification technique are used. Here we are using two types of classification namely SVM and Decision tree. Finally find efficiency of both the algorithms is calculated and it makes to retrieve medical image in large scale at minimum time.

feature space with high density. The different unsupervised algorithms are Feature-Space Based Techniques, Clustering (K-means algorithm, C-means algorithm, Emeans algorithm), Histogram thresholding, Image-Domain or Region Based Techniques(Split-and-merge techniques, Region growing techniques, Neural-network based techniques, Edge Detection Technique), Fuzzy Techniques, etc. Some of the practical applications of image segmentation are:
o o o o o o

IMAGE SEGMENTATION
Segmentation is the process of partitioning an image into multiple segments, also known as super pixels. The goal of segmentation is to simplify and/or change the representation of an image into something that is more meaningful and easier to analyse. Image segmentation is typically used to locate objects and boundaries (lines, curves, etc.) in images. More precisely, image segmentation is the process of assigning a label to every pixel in an image such that pixels with the same label share certain visual characteristics. The result of image segmentation is a set of segments that collectively cover the entire image, or a set of extracted parts from the image. Each of the pixels in a region are similar with respect to some characteristic or computed property, such as or regions are significantly different with respect to the same characteristic(s). Image segmentation algorithms are classified into two types, supervised and unsupervised. Unsupervised algorithms are fully automatic and partition the regions in

Locate tumours and other pathologies Measure tissue volumes Computer-guided surgery Diagnosis Treatment planning Study of anatomical structure

K-MEANS SEGMENTATION In data mining, k-means clustering is a method of cluster analysis which aims to partition n observations into k clusters in which each observation belongs to the cluster with the nearest mean. This results in a partitioning of the data space into Voronoi cells. K-Means clustering generates a specific number of disjoint, flat (nonhierarchical) clusters. It is well suited to generating globular clusters. The K-Means method is numerical, unsupervised, nondeterministic and iterative. The quality of the solution depends on the initial set of clusters and the value of K. The K-Means algorithm can be used to find natural clusters within given data

based upon varying input parameters. Clusters can be formed for images based on pixel intensity, color, texture, location, or some combination of these. K-mean clustering is often suitable for biomedical image segmentation since the number of clusters (k) is usually known for images of particular regions of human anatomy. In biomedical applications, the spatially varying intensity change of a biomedical structure is usually caused by inhomogeneity in the process of image acquisition, such as the inhomogeneous distribution of the contrast agent in CT imaging or inhomogeneous distribution of the magnetic field gradient in MR imaging.

through all the data points results in no data point moving from one cluster to another. At this point the clusters are stable and the clustering process ends. The choice of initial partition can greatly affect the final clusters that result, in terms of inter-cluster and intracluster distances and cohesion. Vertical view Input Image After K-Means Clustering

Horizontal view
Input Image After K-Means Clustering

Properties
There are always K clusters. There is always at least one item in each cluster. The clusters are non-hierarchical and they do not overlap. Every member of a cluster is closer to its cluster than any other cluster because closeness does not always involve the center of clusters.

CLASSIFICATION TECHNIQUE Classification means to assign to each point in the image a tissue class, where the classes are agreed in advance. Generally classification consists of two steps. In first step; a classifier is built describing a pre determined set of data. This step is called as learning step. Next step, the model is used for classification. Goal of classification: previously unseen records should be assigned a class as accurately as possible. In classification, many techniques are available. It is often necessary, therefore, to select a subset of variables before a classifier is generated from data. MRI brain

Process
The dataset is partitioned into K clusters and the data points are randomly assigned to the clusters resulting in clusters that have roughly the same number of data points. For each data point: Calculate the distance (Mahalanobis or Euclidean) from the data point to each cluster. If the data point is closest to its own cluster, leave it where it is. If the data point is not closest to its own cluster, move it into the closest cluster. Repeat the above step until a complete pass

image is classified as Grey matter (GM), White matter (WM), cerebrospinal fluid (CSF). Some of the Classification Techniques are Decision Tree based Methods Rule-based Methods Neural Networks Nave Bayes and Bayesian Belief Networks Support Vector Machine Here image is classified using two classification techniques such as Decision Tree and SVM. And compare which is the best method for classification. DECISION TREE TECHNIQUE The decision trees are most popular of the methods for classification. Decision trees are trees that classify instances by sorting them based on feature values. Each node in a decision tree represents a feature in an instance to be classified, and each branch represents a value that the node can assume. The arcs represent each possible answer to the associated question. Each leaf represents the value of the target variable. Instances are classified starting at the root node and sorted based on their feature values. Advantages of Decision Tree It Can be applied to any type of data. The final structure of the classifier is quite simple and can be stored and handled in a graceful manner. Handles very efficiently conditional information, subdividing the space into sub-spaces that are handled

individually. Reveal normally robust and insensitive to misclassification in the training set. The resulting trees are usually quite understandable and can be easily used to obtain a better understanding of the phenomenon in question. It can be combined with other decision techniques. It is simple to understand and interpret. SVM TECHNIQUE SVM is a popular analysis tool in computer science and engineering. SVM analysis has a fast training and learning speed even when applied to relatively large sample sizes and may be used to evaluate many parameters at once. While SVM analysis can only separate samples into two groups; by using it at each level of decision tree structure, we can separate data into many more specific groups. Support Vector Machines are based on the concept of decision planes that define decision boundaries. A decision plane is one that separates between a set of objects having different class memberships. The basic SVM takes set of input data predicts, for each given input, which of two possible classes form the output. SVM became famous when, using images as input, it gave accuracy comparable to neural-network with hand-designed features in a handwriting recognition task. SVMs can also solve problems with non-linear decision boundaries. Currently the classification of brain image is limited by the image quality. Higher resolution images make glottal area detection both easier and more accurate.

Applications of SVM SVM is widely used in object detection & recognition Content-based image retrieval Text recognition Biometrics Speech recognition From the theoretical perspective that DT can draw straight lines for the classification using information gain algorithm and SVM could draw curve lines. So, logically, SVM could provide better classification for most of the cases. CONCLUSION The proposed system segments MRI brain image using K-means segmentation technique and classifying the incoming 2-D MR image into one of the semantic regions using Decision Tree and SVM algorithm, and gives output as SVM is the best classification method for medical image retrieval and retrieve tumour parts in brain. This technique is used to find diseased brain image of patient to match with database tumour image. Hover, the approach has more general applicability, can use this efficient features and their applications in all image processing techniques. The proposed results show that the performance of the approach is both efficient and effectual for medical data analysis. REFERENCES
1. Yue Gao1, Alicia J. Sprecher2, Jack J. Jiang M.D., Ph.D. Support Vector Machine Based Decision Tree to Classify Voice Pathologies Using High-Speed Videoendoscopy

2. Yanxi Liu, Frank Dellaert and William E. Rothfusa Classification Driven Semantic Based Medical Image Indexing and Retrieval 3. Mahesh pal Paul M.Mather., Decision Tree based classification of remotely sensed data. 4. Robert Burbidge, Bernard Buxton., An Introduction to Support Vector Machines for Data Mining. 5. Tan, Steinbach, Kumar Data Mining Classification: Basic Concepts, Decision Trees, and Model Evaluation. 6. Marcel Prastawa, Guido Gerig Constructing Image Graphs for Segmenting Lesions in Brain MRI 7. Bir Bhanu, Adaptive Image segmentation Using a Genetic Algorithm 8. Caroline Pantofaru A Comparison of Image Segmentation Algorithms 9.Chia-Hao Tsai Yu-Hsiang Wang Segmentation Techniques 10. Jinwei Gu An Introduction of Support Vector Machine. 11. DavidG.Stork Elad Yom-Tov Classification Toolbox. 12.Y. Rui, T.S. Huang, Image retrieval: current techniques, promising directions and open issues, Journal of Visual Communication and Image Representation 10 (1999) 3962. 13.Y. Pang, Yuan Yuan, Xuelong Li, Jing Pan, Efficient HOG human detection, International Journal of Signal Processing 91 (2011) 773 781. 14.A. Vo, Soontorn Oraintara, Nha Nguyen, Vonn distribution of relative phase for statistical image modeling in complex wavelet domain,

International Journal of Signal Processing 91 (2011) 114125. 15.Vincent J. Schmithorst, Marko Wilke. Differences in white matter architecture between musicians and non-musicians: a diffusion tensor imaging. 16.Ahmed Elnakib, Ayman El-Baz, Manuel F. Casanova, Georgy Gimelfarb, and Andrew E. Switala ImageBased Detection Of Corpus Callosum Variability For More Accurate Discrimination Between Dyslexic And Normal Brains".

17.Ahmed Farag, Mostafa Abdelrahman, James Graham, Aly Farag. Shape Modeling of the Corpus Callosum. 32nd Annual International Conference of the IEEE EMBS Buenos Aires, Argentina, August 31 - September 4, 2010. 18.Rong Chen, Edward H Herskovits A Bayesian Network Classifier with Inverse Tree Structure for Voxelwise Magnetic Resonance Image Analysis.

You might also like