Wavelet-entropy based detection of pathological brain in MRI scanning

An accurate diagnosis is important for the medical treatment of patients suffering from brain diseases. Nuclear Magnetic Resonance (NMR) images are commonly used by technicians to assist the pre-clinical diagnosis, rating them by visual evaluations. The classification of NMR images of normal and pathological brains pose a challenge from technological point of view, since NMR imaging generates a large information set that reflects the conditions of the brain. In this work, we present a computer assisted diagnosis method based on a wavelet-entropy of the feature space approach and the Support Vector Machine (SVM) classification method for improving the brain diagnosis accuracy by means of NMR images. The most relevant image feature is selected as the wavelet entropy and is used to train the SVM classifier. The results for over 64 images show that the sensitivity of the classifier is as high as 87.5%, the specificity is 100%, and the overall accuracy is 89.5%. It is easily observed from the data that the proposed classifier can detect abnormality in the brain from the normal controls within excellent performance ranges, which is competitive with latest existing methods.