Texture-based graph regularization process for 2D and 3D ultrasound image segmentation

A Dynamic Graph Cuts Method with Integrated Multiple Feature Maps for Segmenting Kidneys in 2D Ultrasound Images

“…Texture analysis has been adopted to characterize US images in image segmentation studies and led to better performance than image segmentation based on image intensity information alone (Xie, Jiang et al 2005, Faucheux, Olivier et al 2012, Jokar and Pourghassem 2013, Cerrolaza, Safdar et al 2016. Texture feature extraction techniques can be categorized as structural (Mehrnaz and Jagath 2009, Hacihaliloglu, Abugharbieh et al 2011, Jia, Mellon et al 2016, statistical (Wu, Gan et al 2015), model-based (Cohen andCooper 1987, Xie, Jiang et al 2005), and transform-based methods (Haralick, Shanmugam et al 1973, Daugman 1985, Malik, Belongie et al 2001, Cunha, Zhou et al 2006, Jokar and Pourghassem 2013, Cerrolaza, Safdar et al 2016.…”

“…A Gabor transformation consists of a set of filters, which are multiplications of a Gaussian function and a sinusoidal function in different directions and frequencies, capable of capturing texture information similar to perception in the human visual system (Marcelja 1980, Daugman 1985, Jones and Palmer 1987. Although many US image segmentation studies have adopted texture information in the image segmentation (Xie, Jiang et al 2005, Faucheux, Olivier et al 2012, Jokar and Pourghassem 2013, Cerrolaza, Safdar et al 2016, most of them only consider a single texture feature map for image segmentation, which might be insufficient to deal with imaging noise of clinical data.…”