1Detection of Brain Tumor Suspect Areas from CAT scan and MR Images Using CUDA

Abstract—In this paper a solution to identify the tumor suspect areas from the CAT scan and MR images of the human brain is described. Since these images are of very high resolution, a GPU is used for processing the same. GPUs are very efficient at manipulating computer graphics and their highly parallel structure makes them more effective than general-purpose CPUs for a range of complex algorithms. The programming model used in this implementation is the NVIDIA’s CUDA architecture with C as the high level programming language. The proposed algorithm has been tested on a data set of CAT scan and MR images, and the tumor suspect regions were segregated from the respective images. Index Terms—brain tumors, biomedical imaging, mag-netic resonance imaging, image segmentation I