Computer-aided diagnosis for diagnostically challenging breast lesions in DCE-MRI based on image registration and integration of morphologic and dynamic characteristics

Abstract

Diagnostically challenging lesions comprise both foci (small lesions) and non-mass-like enhancing lesions and pose a challenge to current computer-aided diagnosis systems. Motion-based artifacts lead in dynamic contrast-enhanced breast magnetic resonance to diagnostic misinterpretation; therefore, motion compensation represents an important prerequisite to automatic lesion detection and diagnosis. In addition, the extraction of pertinent kinetic and morphologic features as lesion descriptors is an equally important task. In the present paper, we evaluate the performance of a computer-aided diagnosis system consisting of motion correction, lesion segmentation, and feature extraction and classification. We develop a new feature extractor, the radial Krawtchouk moment, which guarantees rotation invariance. Many novel feature extraction techniques are proposed and tested in conjunction with lesion detection. Our simulation results have shown that motion compensation combined with Minkowski functionals and Bayesian classifier can improve lesion detection and classification.

A systematic review on the evaluation and characteristics of computer-aided diagnosis systems

“…Finally, for classification tasks, we observed the use of classifiers such as k-nearest neighbor (Filipczuk et al, 2013;Gedik and Atasoy, 2013;Gopinath and Shanthi, 2013;He et al, 2011;Muramatsu et al, 2013;Nava et al, 2014;Odeh et al, 2006;Osman et al, 2009;Raja et al, 2010;Verikas et al, 2006), artificial neural networks (Barhoumi et al, 2007;Geetha et al, 2008;Jasmine et al, 2009;López et al, 2008;Raja et al, 2007;Streba et al, 2012;Verma, 2009;Wu et al, 2006), Bayesian classifiers (Ampeliotis et al, 2007;Bhooshan et al, 2011;Garnavi et al, 2012;Gruszauskas et al, 2008Gruszauskas et al, , 2009Retter et al, 2013;Tolouee et al, 2011) techniques based on linear discriminant analysis (Lee et al, 2009;Muramatsu et al, 2013;Tanner et al, 2006) and logistic regression models (Shen et al, 2007;Tanner et al, 2006).…”