Enriched Encephalon Tumor Segmentation using Tessellation Automata with High Performed Levelset using Radiosurgery Applications

Segmentation of brain tissues is gaining popularity with the advance of image guided surgical approaches. This work proposes a fast and robust practical tool for segmentation of solid tumors in radiosurgery application. For this, a cellular automata (CA) i. e. Tessellation Structure based seeded tumor segmentation method is used on MR images, which standardizes the volume of interest and seed selection. The procedure starts by establishing the connection of CA based segmentation to the graph theoretic methods to show that iterative CA framework solves the shortest path problem by modifying the state transition function from the CA. A sensitive parameter is introduced to adapt the heterogeneous tumor segmentation problem. Then a smoothness constraint using level set active surfaces is imposed by an EM algorithm over a probability map constructed from CA State. A standard EM level-set propagates normal to its boundary uniformly at a constant speed which will stop on the desired boundary. Finally a CA algorithm is introduced to differentiate necrotic and enhancing tumor tissue with different grade is obtained by a Fuzzy c-means clustering technique to categorize the various type of affected cells content for detailed assessment of radiation therapy response.