Title: A comparison of detrend fluctuation analysis, Gaussian mixture model and artificial neural network performance in the detection of microcalcification from digital mammograms

Authors: Sannasi Chakravarthy S R; Harikumar Rajaguru

Addresses: Department of Electronics and Communication, Bannari Amman Institute of Technology, Sathyamangalam, Tamil Nadu 638 401, India ' Department of Electronics and Communication, Bannari Amman Institute of Technology, Sathyamangalam, Tamil Nadu 638 401, India

Abstract: This paper presents a computer aided approach that classifies the type of cancer (benign or malignant) and its associated risk from the digital mammogram images. Twelve statistical features are extracted through five different wavelets such as Daubechies, Haar, biorthogonal splines, symlets and DMeyer with the decomposition levels of 4 and 6. The Mammogram Image Analysis Society (MIAS) database is utilised in this paper. The micro-calcification in the digital mammogram images is detected by detrend fluctuation analysis (DFA), Gaussian mixture model (GMM) and artificial neural network (ANN). The classifiers' performances are analysed and compared based on the benchmark parameters like sensitivity, selectivity, precision and accuracy. GMM classifier outperforms the DFA and ANN classifiers in terms of performance metrics.

Keywords: mammogram images; breast cancer; wavelet; detrend fluctuation analysis; DFA; Gaussian mixture model; GMM; neural network; classification.

DOI: 10.1504/IJBET.2021.117516

International Journal of Biomedical Engineering and Technology, 2021 Vol.37 No.1, pp.83 - 103

Received: 10 Nov 2017
Accepted: 12 Apr 2018
Published online: 13 Sep 2021 *

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