In health care centers and hospitals, millions of medical images have been generated daily. Analysis has been done manually with an increasing number of images. Brain tumor segmentation in Magnetic resonance imaging (MRI) has been recent area of research in the field of medical diagnosis. Accurate segmentation of brain tumors is an important task and it is challenging problem. K-means clustering algorithm is the most popular and widely-used partitional clustering algorithm in practice. However, traditional k-means algorithm suffers from sensitive initial selection of cluster centers, and it is not easy to specify the number of clusters in advance. Here an Advanced k-means algorithm is proposed for segmentation that can automatically split and merge clusters which incorporate the new ideas in dealing with huge scale of medical image data. Then features are extracted from the segmented image and its efficiency is increased by using Naive Bayes classifier and is classified into normal or abnormal images.