Multilevel learning‐based segmentation of ill‐defined and spiculated masses in mammograms

Abstract

Purpose: A learning-based approach integrating the use of pixel-level statistical modeling and spiculation detection is presented for the segmentation of mammographic masses with ill-defined margins and spiculations. Methods: The algorithm involves a multiphase pixel-level classification, using a comprehensive group of features computed from regional intensity, shape, and textures, to generate a massconditional probability map ͑PM͒. Then, the mass candidate, along with the background clutters consisting of breast fibroglandular and other nonmass tissues, is extracted from the PM by integrating the prior knowledge of shape and location of masses. A multiscale steerable ridge detection algorithm is employed to detect spiculations. Finally, all the object-level findings, including mass candidate, detected spiculations, and clutters, along with the PM, are integrated by graph cuts to generate the final segmentation mask.Results: The method was tested on 54 masses ͑51 malignant and 3 benign͒, all with ill-defined margins and irregular shape or spiculations. The ground truth delineations were provided by five experienced radiologists. Area overlapping ratio of 0.689 ͑Ϯ0.160͒ and 0.540 ͑Ϯ0.164͒ were obtained for segmenting entire mass and margin portion only, respectively. Williams index of area and contour based measurements indicated that the segmentation results of the algorithm agreed well with the radiologists' delineation. Conclusions: The proposed approach could closely delineate the mass body. Most importantly, it is capable of including mass margin and its spicule extensions which are considered as key features for breast lesion analyses.

A spatial fuzzy based level set method for mammogram mass segmentation

“…Tao et. al., [3] proposed an automatic segmentation approach to identify spiculation of a mass. This segmentation algorithm work with the machine learning technique and graph cut algorithm.…”

Mass segmentation in mammograms using a kernel-based fuzzy level set method

“…Many algorithms have been developed in the area of mammogram mass segmentation (Huo et al, 1995;Guliato et al, 1998;Kom et al, 2007;Yading et al, 2007;Domínguez and Nandi, 2008;Shi et al, 2008;Song et al, 2009;Dubey et al, 2010;Elter et al, 2010;Song et al, 2010;Tao et al, 2010;Liu et al, 2011;Kai et al, 2011;Pang et al, 2012;Xin et al, 2012;Berber et al, 2013;Danilo et al, 2014;Huang et al, 2014), such as thresholding methods, clustering methods, region-based methods and edge-based methods. Dominguez and Nandi (2008) introduced a thresholding approach that uses combination of different thresholding levels to detect masses.…”

“…A multiple thresholding method is introduced by Danilo et al (2014) which combine wavelet analysis and genetic algorithm for computing optimal threshold values to detect and segment the masses. Tao et al (2010) proposed an automatic segmentation approach to identify speculation of a mass. Tao et al (2010) proposed an automatic segmentation approach to identify speculation of a mass.…”

A Novel Breast mass segmentation method based on patch merging and GHFCM

“…In [13], Kobatake, et al, applied a modified Hough transform to extract lines passing near the centre of the mass and automatically selected candidates based on the number of line-skeletons. In [15], Elter, et al, proposed a contour tracing approach to extract shape of the region. The ROI of mass is transformed into polar coordinate system, and then contour of estimated mass is calculated by using a shortest path algorithm.…”