F1: Clinical Engineering

SVM CLASSIFIER OF MRI IMAGES FOR COMPUTER-ASSISTED DIAGNOSIS

Madina Hamiane, Fatema Saeed

Ahlia University, Bahrain

Magnetic Resonance Imaging is a powerful technique that helps in the diagnosis of various medical conditions. Detection of brain abnormalities, such as brain tumors, in brain MRI images are considered in this work. These images are often corrupted by noise from various sources. The images are first pre-processed using the Discrete Wavelet Transforms (DWT) along with thresholding techniques for efficient noise removal. Edge detection and threshold segmentation are next applied to the denoised images prior to the extraction of the segmented image features through the use of morphological operations. The images are finally classified using an improved Support Vector scheme. The results obtained in the pre-processing stage show that biorthogonal 1.3 wavelet of the first level DWT and hard threshold gave the minimum MSE, highest SNR and highest PSNR which were selected as performance metrics.  Morphological operations are  used to extract the region of interest whose features are subsequently calculated from the Gray Level Cooccurrence Matrix . These features are used to train an improved Support Vector Machine classifier that uses a Gausssian radial basis function kernel . The performance of the classifier is evaluated and the the results of the classification show that the proposed scheme accurately distinguishes normal brain images from the abnormal ones which are further efficiiently classified as exhibiting begin or malignant tumors. The accuracy of the classification is shown to be as high as 99% which is superior to the results reported in the literature.

The proposed accurate automatic classification of the SVM classifier can be used by neurologists to help them identify brain abnormalities that might be hidden due to the large number of slices that are obtained from MRI brain images.
 

Organised by

Endorsed by

 

        Supported by
 

 

                          
     

       

                             
           
 

        Supporting Media