Detection of brain tumor using modified particle swarm optimization (MPSO) segmentation via haralick features extraction and subsequent classification by KNN algorithm. (2022)

• Record Type: Journal Article
• Authors: Deepa, G.; Mary, G. Leena Rosalind; Karthikeyan, A.; Rajalakshmi, P.; Hemavathi, K.; Dharanisri, M.
• Is Part Of: Materials today. Volume 56:Part 4(2022)
• Page Start: 1820
• Page End: 1826
• Publication Date: 2022
• Subjects: Modified particle swam optimization -- Haralic feature -- KNN algorithm -- Brain tumor -- Image processing; Materials science -- Congresses -- Periodicals; 620.1
• DOI: 10.1016/j.matpr.2021.10.475
• Languages: English
• ISSNs: 2214-7853

Abstract

Abstract: Brain tumors used to be uncommon, but they are increasingly becoming more common. Magnetic Resonance Images (MRI) scans may also be used by the medical specialist to detect brain malignancies. One of the most critical concerns in artificial intelligence systems is medical diagnosis via image processing and machine learning. Artificial intelligence based machine learning has enabled the digitalized the process of health monitoring. Diagnostic Images are a valuable source of information for determining the prognosis and treatment of certain cancers. It takes shape when human cells do not develop, split, or perish in the same way as conventional cells do. All voluntary and involuntary actions in the human body are controlled by the brain. Maintaining a healthy brain is critical for living a longer life. This paper proposes brain tumor detection using modified particle swarm optimization (MPSO) segmentation with ensemble classification. Initially image noise is suppressed using wiener filter. Then the feature extraction process is implemented using Haralick features. The SVM classifier with improved fuzzy segmentation is compared to the proposed work. With an accuracy of 98.2 percent, the new model outperformed the previous model.