Vascularity-Oriented Level Set Algorithm for Vessel Segmentation in Medical Image Processing
|School||Shanghai Jiaotong University|
|Keywords||Vessel segmentation Active contour models Level set algorithm Vascular feature extraction Three-dimensional medical image Image Auxiliary Diagnosis and Treatment|
In medical image processing , 3D visualization of blood vessels with a wide range of applications . It not only can help to understand the morphology of vascular physiology , anatomy and blood flow state , but also play an important role in the diagnosis of vascular disease . In addition , auxiliary clinic platform in medical image the precise surgical path planning has a crucial role in the success of the surgery . For example , in the diagnosis and treatment of lung cancer intrusive , accurate pulmonary vascular segmentation for preoperative planning and surgical navigation image visualization conditions , in order to avoid the destruction of the main blood vessels and cause unnecessary harm to patients . Although simple threshold segmentation algorithm to distinguish different gray-scale range of organizations , accurate results and can not be used in surgery. In recent years, the level set algorithm for active contour models have been widely and successfully applied to the segmentation of medical images in a variety of tissues and organs . However , the traditional algorithm , the inhibitory effect of high curvature reduces the speed of evolution of the evolution of the curve in the direction along the vessel , thus creating difficulties vessel segmentation . This paper presents a new level set algorithm based on tubular features , eliminating the impediment to high curvature , guide curves along the vessel direction in the evolution of travels faster . At the same time , the probability density function to determine the level set curve according to the distribution of blood vessels and the background point to the direction of the expansion or contraction of evolution . The new algorithm in the actual vascular CT images Experimental results demonstrate that it can quickly and effectively dividing the three-dimensional blood vessel .