Medical image registration of the brain

Abstract

Image processing methods possibly able to visualize objects inside the newlinehuman body, are of special interest. Advances in computer science have led newlineto reliable and efficient image processing methods useful in medical newlinediagnosis, treatment planning and medical research. In clinical diagnosis newlineusing medical images, segmentation of objects is a critical task due to noise newlineand multi-resolution of objects. The better segmentation can be performed newlineby the integration of useful data obtained from separate images. The images newlineneed to be geometrically aligned for better observation. This procedure of newlinemapping points from one image to corresponding points in another image is newlinecalled Image Registration. It is a spatial transform. It is used in computer newlinevision, medical imaging, military automatic target recognition and newlinecompiling and analyzing images and data from satellites. newlineIn this dissertation work Image Registration is used for the newlinesegmentation of brain tumors. Currently, cancer is one of the important newlinediseases in the world; both developed and developing nations are suffering newlinedue to different causes. In human body, brain is an important organ, newlinecontrols the function of all other organs. In medical images once object is newlinesegmented, the diagnosis and treatment planning can be easily performed newlineby the doctors based on the nature an d growth rate of tumors. In this work, newlineactive contour segmentation of brain tumors is performed with the aid of newlineregistration. newlineImage registration uses two images. The reference and the sensed newlineimage could be different because they are taken at different times(multi newlinetemporal), using different devices like Magnetic Resonance Imaging(MRI), newlineComputer Tomography(CT), Positron Emission Tomography(PET), Single newlineProton Emission Tomography(SPECT)etc. (multi modal)and from different newlineangles (multi-view)in order to have 2D or 3D perspective. Image newlineregistration is the process of transforming different sets of data into one newlinecoordinate system. Registration is necessary in order to be able to compare newlineor integrate the data obta