TY - JOUR AU - Murthi, A. AU - Shameer, S. PY - 2016/12/26 Y2 - 2024/03/28 TI - A Novel Two-Stage Approach For Automatic Detection of Brain Tumor JF - JOURNAL OF ADVANCES IN CHEMISTRY JA - JAC VL - 12 IS - 25 SE - Articles DO - 10.24297/jac.v12i25.4427 UR - https://rajpub.com/index.php/jac/article/view/4427 SP - 5653-5660 AB - <span lang="X-NONE">Brain tumor is one of the most life-threatening diseases, and it is the <span>most common type of cancer that occurs among those in the age group belonging to 0-19. It is also a major cause of cancer-related deaths in children (males and females) under age 20</span> hence its detection should be fast and accurate. Manual detection of brain tumors using MRI scan images is effective but time</span><span lang="EN-IN">-</span><span lang="X-NONE">consuming. Many automation techniques and algorithms for detection of brain tumors are being proposed recently. In this paper, we propose an integrated two-step approach combining modified K-means clustering algorithm and Hierarchical Centroid Shape Descriptor (HCSD). The images are clustered using modified K-means based on pixel intensity, and then HCSD helps to select those having a specific shape thus making this approach more effective and reliable. Simulation of the proposed work is done in MATLAB R2013a. Tests are carried out on T1 weighted MRI scan images.</span> ER -