A graph-based segmentation method for 3D ultrasound images

As the 3D imaging techniques have been more and more widely applied in medical applications, analysis of 3D medical images becomes necessary. This paper presents a novel segmentation method for extracting objects of interest (OOI) in 3D ultrasound images. The proposed method taking advantage of graph theory to construct a 3D graph is called 3D graph-based segmentation algorithm. It can generate a set of minimum spanning trees each of which corresponds to a 3D sub-region. In comparison with a previously reported 3D active contour model (i.e. 3D Snake), our graph-based segmentation method is less computationally complex, hence resulting in less computation time. In terms of segmentation accuracy, the experimental results using an ultrasound phantom demonstrate that our method outperforms the 3D Snake and Fuzzy C Means clustering methods, indicating improved performance for potential clinical applications.