Superpixel-based segmentation and automated identification of active tumour and necrotic regions in bone tumor using t1 and diffusion weighted imaging.

Abstract

Proper Delineation of the tumour boundary and assessment of tumour size can take crucial part in treatment planning and monitoring treatment response. We investigate a fully automated Simple linear iterative clustering (SLIC) superpixel-based method for detection and segmentation of pathological tissues like oedema, tumour and necrosis associated with Osteosarcoma. Experimental results provide a close match to expert delineation and was able to estimate areas of active tumor and necrosis with good accuracy.