Current Trends in Segmentation of Medical Ultrasound B-mode Images: A Review

Though the classical image-segmentation techniques have demonstrated their capability in various -applications, their performance degrades severely when they are applied to segment the still Ultrasound (US) images owing to the speckle, tissue related textures, and the artefacts resulting from the ultrasonic imaging process. Although these three types of signals make a clinical decision difficult, the deterioration of edge definition caused by them severely degrades the performance of most classical boundary - detection algorithms. This review gives a bibliographical survey of the current research efforts in the area of image segmentation of still US medical images and presents some of the challenges anticipated in the years to come. A full description of the various methods is beyond the scope of this article; instead, the focus is on providing an introduction to the methods of segmentation.