Title: Detection of brain tumour by using moments and transforms on segmented magnetic resonance brain images

Authors: Ajay Prashar; Rahul Upneja

Addresses: Sri Guru Granth Sahib World University, Fatehgarh Sahib, Punjab, India; Department of Mathematics, Trinity College, Jalandhar, Punjab, India ' Department of Mathematics, Sri Guru Granth Sahib World University, Fatehgarh Sahib, India; Department of Electrical and Computer Engineering, University of Manitoba, Winnipeg, Canada

Abstract: Brain tumour occurs when abnormal cells appear within the brain. Primary tumour starts with abnormal growth of brain cells whereas Secondary (Metastatic) tumour initiates as cancer in other parts of the body and spread to the brain through blood stream. In this paper, we propose a novel approach to detect tumour in magnetic resonance (MR) brain images. The proposed method uses improved incremental self organise mapping (I2SOM) to segment the brain image and to calculate asymmetry Zernike moments (ZMs), Pseudo-Zernike moments (PZMs) and orthogonal Fourier Mellin moments (OFMMs) are used. It omits the limitation of pre-determination of class of input data and the manual setting of appropriate threshold value. The effectiveness of the proposed method is analysed by doing experiments on 30 MR brain images with tumour and 30 normal MR brain images. It is observed that tumour detection is successfully realised for 30 MR brain images with tumour.

Keywords: tumour detection; Zernike moments; ZMs; Pseudo-Zernike moments; PZMs; orthogonal Fourier Mellin moments; OFMMs; polar harmonic transforms; segmentation.

DOI: 10.1504/IJCSM.2020.111109

International Journal of Computing Science and Mathematics, 2020 Vol.12 No.2, pp.157 - 176

Received: 19 Oct 2017
Accepted: 21 Nov 2017
Published online: 10 Nov 2020 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article