MRI Image Processing Using Different Techniques

Authors: Rida Khan, Mrs.Nisha Bhati

Certificate: View Certificate

Abstract

A Wavelet Transform based image decomposition algorithm is proposed to identify the areas of interest in the brain related to in oxidation problems. The significant brain activations can be observed in Magnetic Response Imaging (Images) & EEG signals related various brain functions/ disorders. The image de-noising using wavelet transform and subsequently feature extraction for classification is the way of many researchers for classification of medical images. In this paper, the study of the brain feature extraction in MRI images using various techniques has been given. The different classifier has been reported in literature for the improvement of classification performance. The comparison of the different classification methods of the medical image database is also given in the paper.

Introduction

Magnetic Resonance Imaging (MRI) has become a widely employed high quality medical imaging nowadays in the field of tumor detection. Brain tissue and tumor segmentation in MR images have become a vital area of discussion. For accurate image segmentation, some good features have to be extracted. The brain is comprised of different tissues such as the White Matter (WM), Cerebrospinal Fluid (CSF) and Gray Matter (GM). During the segmentation of the MR brain images, variability in certain aspects such as, tumor shape, location, size, intensity and textural properties makes the segmentation process difficult. In tumor segmentation, intensity feature plays a vital role in differentiating tumor from other brain soft tissues. But, intensity alone is not sufficient, therefore other texture based features such as Local Binary Pattern (LBP), gray level based features, Gray Level Cooccurrence Matrix (GLCM), wavelet features are extracted This is medical imaging techniques analysis tools enable both doctors and radiologists to arrive at a specific diagnosis process. Medical Image Processing has emerged as one of the most important tools to identify as well as diagnose various disorders. Imaging helps of the doctors to visualize and analyze the image for understanding of abnormalities in internal structures. In brain images data obtained from Bio-medical Devices which use imaging techniques like Magnetic Resonance Imaging (MRI), Computed Tomography (CT) and mammogram function, which indicates the presence or absence of the lesion along with the patient history, of important factor in the diagnosis process. Magnetic Resonance Imaging (MRI) is a scanning device that uses magnetic fields and computers to capture images of the brain on film (brain, image). It does not use x-rays. The provides pictures from various planes and process, which permit doctors to create a threedimensional image of the tumor. MRI are brain image detects signals emitted from normal and abnormal tissue detects, and providing most tumors images. The become a widely-used method of high quality medical image, brain imaging where soft tissue contrast and non-invasiveness are clear advantages. Brain images have been selected for the image reference for this research because the injuries to the brain tend to affect large areas of the organ process. The brain controls and coordinates most movement, homeostatic body functions such as heartbeat and behavior, fluid balance, body temperature, or blood pressure. In the functions of the brain are responsible in different category for example cognition, memory, motor learning and other sorts of learning, emotion. The classifications of brain MRI data as normal and abnormal are important to prune the normal patient and to consider only those who have the possibility of having abnormalities or tumor.

Conclusion

The various papers and literature has been studied for MRI image classification. The comparisons of the methods have been given in the form of table. The Support Vector Machines (SVM) perform better in the recognition but required higher computation time. In feature, the Neural network and fuzzy classification with various other feature extraction techniques may be useful. The parallel processing may also the option to use the multi-core processer to reduce the time of computation.