Improved Brain Tumor Detection Technique Using Hybrid Roundness Metric, Region Growing and Cellular Automata Edge Detection

Abstract: Brain tumor is one of the most life scowling diseases which need urgent and accurate detection. Various automated techniques have been designed to detect the tumor efficiently from MRI imaging. The aim of this paper is to propose a new and improved method for increasing the efficiency of tumor detection. The proposed technique involves the integration of Hybrid Roundness Metric with the existing techniques of segmentation called Modified Texture Based Region Growing and Cellular Automata Edge Detection. Hybrid Roundness Metric detects the circular objects thereby easing the detection of exact location of tumor. The two segmentation techniques perform well in collaboration but when Hybrid Roundness Metric is implemented along with the above two, results achieved are much more effectual. The detection of tumor is made easier and accurate with the proposed work.

Keywords: MRI; hybrid roundness metric; region growing; edge detection; tumor detection improved efficiency, parameters.