Publication Type : Book Chapter
Publisher : Elsevier
Source : Deep Learning Techniques for Biomedical and Health Informatics
Url : https://doi.org/10.1016/b978-0-12-819061-6.00004-5
Campus : Amaravati
School : School of Computing
Department : Computer Science and Engineering
Year : 2020
Abstract : In this chapter, we propose a novel approach called the HARIS-Heuristic approach for real-time segmentation to analyze a brain's magnetic resonance image (MRI) for tumor identification. MRIs are pre-processed using an adaptive contourlet transform to remove noise and a structural augmentation technique to remove the skull region. Resultant MRIs are fed as an input to HARIS, which segments them based on texture using multiobjective functions. To evaluate the efficiency of HARIS, a comparison is made with existing techniques such as Twin Centric GA-SGO, CNN, and GA-TLBO. HARIS outperforms its counterparts in accurately identifying tumors with the least computational time.
Cite this Research Publication : P. Naga Srinivasu, T. Srinivasa Rao, Valentina Emilia Balas, A systematic approach for identification of tumor regions in the human brain through HARIS algorithm, Deep Learning Techniques for Biomedical and Health Informatics, Elsevier, 2020, https://doi.org/10.1016/b978-0-12-819061-6.00004-5