Segmentation of brain tumours for radiosurgery applications using image processing

Abstract

Background and Aim: The proposed work is to segment the solid tumours with user interaction to assist researchers in
Radiosurgery planning. The brain tumour segmentation methods rely on the intensity enhancement. Methods: In this work,
Cellular Automaton (CA) based seeded tumour segmentation algorithm is proposed. Which determine the Volume of Interest
(VOI) and seed selection is done based on the user interaction. Results: First, establish the connection of the CA-based
segmentation to the Tumour-cut method to show that the iterative CA framework solves the shortest path complication. In that
regard, the proposed method modify the state transition function of the CA to calculate the shortest path solution. Furthermore, an
algorithm based on CA is presented to differentiate necrotic and enhancing tumour tissue content, which gains importance for a
researcher in planning therapy response. The tumour-cut algorithm run twice for background seed (healthy cell) and foreground
seed (tumour cell) for probability calculation. Among them, a clustering method have been investigated and used. Conclusion:
Finally, this paper applied Tumour-Cut method and K-means clustering to differentiate necrotic and enhancing tumour tissue
content, which gains importance for a complete evaluation.