Brain Tumor Classification Based on Neural Network and Region Growing Segmentation
Image classification recently becomes an essential high level process after segmentation in image processing. Image segmentation is the process by which the original image is subdivided into its constituent regions or objects. Image segmentation is an initial task for higher level image processing such as classification or objet recognition. In medical image segmentation, structures or objects of interest for segmentation include abnormalities like brain tumor. Detecting brain tumor in human brain using magnetic resonance (MR) image is important in medical imaging for diagnosis of the disease. Automatic detection of brain tumor in MRI provides the abnormal tissue which is necessary for treatment planning. The exact boundary tumor detect by using the image segmentation. In this paper we perform an objective comparision of region based segmentation techniques and classification of tumor. The performance of the segmentation algorithm are evaluated by using the methods Rand Index(RI), Variation of information (VOI) and Global Consistency Error (GCE) for MR image. Artificial neural network (ANN) is used to classify mainly in medical imaging. Radial Basis Function (RBF) and feed forward back propagation (FFBP) are trained the feature vector for classification. Features are calculated from the extracted image.
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.