A computational early vision model for segmentation of clinical ultrasound images

Image segmentation is a fundamental step for quantitative ultrasound image analysis. However, due to the intrinsic noisy nature of an ultrasound image, classic segmentation techniques are usually ineffective in performing ultrasound image segmentation. Here, the authors present a computational early vision model for segmentation of clinical ultrasound images. Their approach is based on computing perceptual similarity among local blocks of images, which has been shown to be promising by experimental results on real ultrasound images