摘要
血管内超声(IVUS)图像的冠状动脉血管壁内、外膜边缘提取对冠脉疾病的诊断和治疗有着重要意义。针对实际IVUS图像血液斑点噪声比较严重的情况,提出一种基于超声序列图像斑点噪声抑制和活动轮廓模型(Snake模型)的IVUS图像边缘提取方法。首先采用一种时/空滤波方法对IVUS图像进行降噪预处理,该方法能够有效地抑制IVUS图像的血液斑点噪声;然后基于Snake模型和图像的统计特征自动提取冠脉血管壁内、外膜边缘。实验结果表明,本算法简单,准确性较高,对序列图像处理的可重复性和鲁棒性较强,是一种较好的全局最优化算法。
The intima and adventitia edge detection of coronary artery of intravascular ultrasound (IVUS) images is important in the diagnosis and treatment of the coronary artery disease. Due to the prominent blood speckle noise in real IVUS images, a new method based on speckle noise reduction of ultrasound image sequences and the snake model was proposed for detecting automatically the edge oflVUS images. First, a temporal/spatial filtering method which can suppress speckle noise significantly was used to preprocess the IVUS images. Then, the intima and adventitia edge is detected automatically based on the active contour model and statistical features of IVUS images. Experiments show that the method is algorithmically simple, statistically accurate, reproducible and robust in sequential IVUS frames, and it is a kind of global ootimal algorithm.
出处
《系统仿真学报》
EI
CAS
CSCD
北大核心
2006年第7期1999-2002,2045,共5页
Journal of System Simulation
基金
国家自然科学基金资助项目(60271015)
山东省优秀中青年科学家科研奖励基金项目(2005BS01006)。
关键词
血管内超声
斑点噪声
时/空滤波
边缘提取
活动轮廓模型
动态规划
intravascular ultrasound
speckle noise
temporal/spatial filtering
edge detection
active contour model
dynamic programming