Abstract

Brain tumor detection is a very challenging task in today’s medical world. Nowadays many doctors prefer ready-made methods to diagnose tumorsby examiningmagnetic resonance imaging (MRI) images. Accurate and less time-consuming methods are used frequently to identify tumors. This chapter proposes a couple of methods for identification and classification of brain tumors. The proposed methods use support vector machine (SVM) and convolution neural network (CNN) with Softmax classifier. The SVM uses histogram of oriented gradients (HOG) as a feature to label the given dataset. Further, the same feature of the test image is verified against labeled sets and classified. The SVM gives an accuracy of 87.33%. In CNN, a multi-layer structure is used. This contains a convolution layer followed by a pooling layer and, finally, a fully connected layer. In CNN, Alexnet is preferred for classification. In Alexnet, the rectified linear unit (ReLU) is used as an activation function. Following its output, Softmax is used as an activation to classify the images. The CNN with Softmax classifier gives an improved accuracy of 98.68%. In addition, the performance measures, such as specificity, sensitivity, and precision, are also determined. The sensitivity with SVM is 93.18%, while with CNN it increases up to 100%. Comparing both methods with the obtained results, the CNN with Softmax classifier gives better results. The proposed method with CNN produces only one misclassified image, offering an advantage over the existing method, while SVM generates nine misclassified images. The methods are tested on 253 MRI images that have been detected and classified correctly with Matlab.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call