IJRET: International Journal of Research in Engineering and Technology eISSN: 2319-1163 | pISSN: 2321-7308 __________________________________________________________________________________________ Volume: 03 Special Issue: 07 | May-2014, Available @ http://www.ijret.org 470 PATTERN –BASED WITH SURFACE-BASED MORPHOMETRY SURVEY ON BRAIN CHANGES R.Keerthana 1 , M.Babu 2 , S.R.Sridhar 3 1 PG Scholar (M.E), Department of Computer science, Knowledge Institute of Technology, Salem, Tamil Nadu, India 2 Assistant Professor, Department of Computer science, Knowledge Institute of Technology, Salem, Tamil Nadu, India 3 Assistant Professor, Department of Computer science, Salem, Tamil Nadu, India Abstract Morphometry is identifying and characterizing differences and correlations between brain shapes among population. Study of brain shape has drawn attention among many researchers on different diseases counting Schizophrenia, dyslexia, autism, Alzheimer and turner’s syndrome. Many approaches have been proposed for computer-assisted diagnostic taxonomy. Several significant progressive brain changes occur during aging. Pattern-Based morphometry is a robust application for measuring the change in brain parametric mapping. By reason of bias featuring in image along with the algorithm, have a penalty term and inverse consistency that are necessary to oversight the change in non-biological structure. Morphometric analysis forces a pact between the sensitivity and specificity that has drawn reporting or increasing attention in the field of biological science. A novel technique illustrating the brain tissue aging survey with surface-based and multivariate pattern of morphometric brain change To keep the robustness and specificity contributed by the spatial term and cortical analyses, while maintaining the localization and sensitivity Experimental results propose a greater inter variability within normal aging as well as the generation of more sensitive based morphometry in brain survey. Keywords: Image registration, Multivariate analysis, Cortical analysis, Pattern based morphometry (PBM), Surface based Morphometry (SBM). ------------------------------------------------------------------------***---------------------------------------------------------------------- 1. INTRODUCTION Medical Imaging with Biomedical Engineering has become an important component in our day to day clinical applications .A survey on detecting change in non –biological structure is a challenging task which can be either longitudinal or cross- scale[4]. Estimations of survey are dependent by many factors .Some factors are with the images and others with its methods. There are different forms of analysis of brain such as Single- photon emission computed tomography (SPECT), Computed tomography (CT), positron Emission Tomography (PET), and Magnetic resonance imaging (MRI) ,in which it is widely used for analysis. These are used to detect abnormality localization in patients. They are subjected to noise while segmentation. The major difficulty in brain change detection is to model change in terms of deformation at region boundaries from the boundary-based method. Morphometry of whole brain to detect abnormalities has approaches like Voxel-based to quantify mass-univariate or multivariate analyses and Tensor-Based Morphometry (TBM) [1] obtained from registration to measure the amount of deformation to measure the anatomical features of the subject. The TBM deals with the sensitivity and specificity with reference to the brain changes. Deformation varies in different ways (i.e.) the amount of deformation depends on the morphology of the subject. Fig-1 Six regions present in brain image usually used for morphometric analysis. In this paper we will exploit a new technique Pattern-Based Morphometry (PBM) which a data driven technique. It uses sparse dictionary learning algorithm that is used to identify the differences globally. This is done along with another method Surface-based Morphometry (SBM) [4] which measures from geometric models of cortical surfaces. 2. RELATED WORK The specificity of region-based approach along with the boundary-based approach were used to combine the localization and sensitivity advantages. By using TBM we can find the volume changes typically that appear at tissue