Brain Tumor Detection and Classification Via Kernel Support Vector Machine Using MRI Images

Abstract

The MR images of the brain requires automatic and accurate classification for medical analysis and interpretation nowadays. Numerous methods have been declared already in the previous years. In this paper we have presented a method which classifies the brain image of MRI into normal and abnormal brain tumor images. This method uses wavelet transform that extract features from images. The next step involves principle component analysis (PCA) that reduces the feature dimensions. These reduced features are then employed to a kernel support vector machine (KSVM). 180 images of the brain were collected from the diseased which contained 130 abnormal brain and 50 normal brain images. Four kernels of different types were performed.

Introduction

I. INTRODUCTION

To produce high quality images of the anatomical structure of the human body, especially in the brain, Magnetic resonance imaging (MRI) is used. It provides clear information of anatomical parts for clinical diagnosis and biomedical research.  For feature extraction from magnetic resonance brain images 2D discrete wavelet transform is the most effective tool. Because that allows analysis of images at various levels of resolution. It is due to the multi-resolution analytic property. In this principle component analysis is used to reduce the feature vector dimensions and to increase the discriminative power.

Classification of the image is the main purpose of this work. The brain magnetic resonance images are classified using Support vector machine and K-nearest neighbor methods which are supervised classification method. In unsupervised classification method fuzzy c-means algorithm is used. However, the classification accuracies of most of the existing methods were lower than 95%, so the goal of this paper is to find a more accurate method of classification.

In this paper, we used Kernel support vector machine (KSVM) which allows to fit the maximum-margin hyperplane in a transformed feature space.

Compared with other methods such as decision tree, Bayesian network and artificial neural network, KSVMs have significant advantages of high accuracy direct geometric interpolation etc.

II. METHODOLODY

A. Database

The datasets consists of T2-weighted MR brain image and 256 × 256 in-plane resolution, which were downloaded from the website of OASIS dataset (URL: http:// www.oasisbrains.org/). T2 model is chosen since T2 images are of higher-contrast and clearer vision compared to T1 and PET modalities.

B. Preprocessing

The first step in preprocessing is the Image resize. Image resizing is a process of changing pixel information, resizing an image involves changing the size of the pixels without cutting anything out. Image resizing to regulate the images to fixed scale (512×512) in order that it supports the classification with clear and accurate features. After that, conversion of images from RGB to gray level was done.

The next step is morphological operation of the image. Morphology is a set of image processing operations that process images based on size and shapes. Morphological operations applies a structuring element to an input image which creates an output image of the same size. In this step, processes like erosion, dilation and inversion are done.

Conclusion

Brain tumors are caused by abnormal and uncontrolled growing of the cells inside the brain. Treatment of a brain tumor depends on its size and location. Although benign tumors do not tend to spread widely, they can cause damage by pressing on areas of the brain if they are not treated early. To avoid manual errors in the classification process, an automated intelligent classification technique is proposed which caters the need for classification of image. In this proposed work classification techniques based on Support Vector Machines (SVM) are proposed and applied to brain image classification. Here the proposed brain tumor image segmentation based on discrete wavelet transform (DWT). The proposed work is tested with SVM classifier models. This automated intelligent system will results in the improvement of accuracy and reduces the error of MRI brain tumor.

References

[1] Birare, Geetanjali, and V. A. Chakkarwar, \"Automated detection of brain tumor cells using support vector machine\", 9th International Conference on Computing, Communication and Networking Technologies (ICCCNT), pp. 1-4. IEEE, 2018. [2] Srinivas, B., and G. Sasibhushana Rao, \"Performance evaluation of fuzzy C means segmentation and support vector machine classification for MRI brain tumor\", Soft computing for problem solving, pp. 355-367. Springer, Singapore, 2019. [3] Mishra, S.K. and Deepthi, V.H., “Brain image classification by the combination of different wavelet transforms and support vector machine classification”, Journal of Ambient Intelligence and Humanized Computing, Vol. 12, No.(6), pp.6741-6749, 2021. [4] Rampisela, T.V. and Rustam, Z., “Classification of schizophrenia data using support vector machine (SVM)”, Journal of Physics: Conference Series, Vol. 1108, No. 1, pp. 012044, IOP Publishing, 2018. [5] Liu, J., Xu, H., Chen, Q., Zhang, T., Sheng, W., Huang, Q., Song, J., Huang, D., Lan, L., Li, Y. and Chen, W., “Prediction of hematoma expansion in spontaneous intracerebral hemorrhage using support vector machine”, EBioMedicine, 43, pp.454-459, 2019. [6] Gurbin?, Mircea, Mihaela Lascu, and Dan Lascu, \"Tumor detection and classification of MRI brain image using different wavelet transforms and support vector machines\", 42nd International Conference on Telecommunications and Signal Processing (TSP), pp. 505-508. IEEE, 2019. [7] Hossain, T., Shishir, F.S., Ashraf, M., Al Nasim, M.A. and Shah, F.M., “Brain tumor detection using convolutional neural network”, 1st international conference on advances in science, engineering and robotics technology (ICASERT), pp. 1-6, IEEE, 2019. [8] Gumaste, P.P. and Bairagi, V.K., “A hybrid method for brain tumor detection using advanced textural feature extraction”, Biomedical and Pharmacology Journal, Vol.13, No.(1), pp.145-157, 2019. [9] Devkota, B., Alsadoon, A., Prasad, P.W.C., Singh, A.K. and Elchouemi, A., “Image segmentation for early stage brain tumor detection using mathematical morphological reconstruction”, Procedia Computer Science, Vol. 125, pp.115-123, 2018. [10] Ejaz, K., Rahim, M.S.M., Rehman, A., Chaudhry, H., Saba, T., Ejaz, A. and Ej, C.F., “Segmentation method for pathological brain tumor and accurate detection using MRI”, International Journal of Advanced Computer Science and Applications, Vol. 9, No.(8), pp.394-401, 2018.

Copyright

Copyright © 2024 Ms. P. Divya Jenifar, Ms. R. Keerthana. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.