Implementation of Vertical Search technique for diagnosis of Abnormal Tumor Region from Brain CT Images
M. Ragavi, T. L. Nija, Shining Gold · 2014
Abstract — Automatic segmentation of brain computed tomography images is an important task. Automating this process is challenging due to high diversity in appearance of tumor tissue among different patients and in many cases, similarity between tumor and normal tissue. After the classification and segmentation of a brain computed tomography images vertical search is implemented on the images. Diagnosis of tumor region from the brain computed tomography image can be done by using vertical search implementation. A Dominant gray level run length texture feature set is derived from the ROI of the image is to be selected. We construct the SVM based classifier and evaluate the performance by comparing the classification results.To improve the computing efficiency it select the most suitable feature extraction method that can be used for classification and segmentation of brain tumor in computed tomography images efficiently and accurately. Vertical search method entirely searches the segmented region space.