李锵, 张琦珺, 关欣, 滕建辅
AuthorsHTML:李锵, 张琦珺, 关欣, 滕建辅
AuthorsListE:Li Qiang, Zhang Qijun, Guan Xin, Teng Jianfu
AuthorsHTMLE:Li Qiang, Zhang Qijun, Guan Xin, Teng Jianfu
Unit:天津大学微电子学院，天津 300072
Unit_EngLish:School of Microelectronics, Tianjin University, Tianjin 300072, China
Abstract_Chinese:颈动脉的内中膜厚度(IMT)是预测心血管疾病(CVDs)病发程度的重要指标．本文研究并提出一种自动、高效的计算机辅助IMT测量算法, 该算法依据先验知识自动提取感兴趣区域(ROI), 并采用基于隐马尔可夫随机场(HMRF)模型改进的模糊C均值(FCM)算法分割图像, 实现IMT的自动测量．实验结果表明, 所提算法对超声图像噪声的鲁棒性较强, IMT自动测量结果与真实值(GT)有很高的一致性：两个数据集合的相关系数为98.52%, 平均绝对误差为

.
Abstract_English:Common carotid artery intima-media thickness(IMT)，which is considered as an important indicator of the development of cardiovascular diseases(CVDs)，is usually measured on ultrasound images. This paper proposed an automatic and efficient computer-aided IMT measurement system. With the proposed method，region of interest(ROI)is extracted automatically based on prior knowledge，and then improved fuzzy C means(FCM)based on hidden Markov random field(HMRF)were applied to segment ROI and to detect the boundary of intima-media for IMT measurement. Experimental results show that the proposed method has strong robustness against ultrasound artifacts，and achieves similar clinical parameters to the ground truth(GT). The correlation coefficient between the two databases reaches 98.52% ，and the mean of the absolute error between them is
.
Keyword_Chinese:超声图像分割; 内中膜厚度测量; 模糊C均值; 隐马尔可夫随机场模型; 感兴趣区域
Keywords_English:ultrasound image segmentation; intima-media thickness(IMT)measurement; fuzzy C means(FCM); hidden Markov random field(HMRF)model; region of interest(ROI)

PDF全文下载地址:http://xbzrb.tju.edu.cn/#/digest?ArticleID=5972