百分比白:颈动脉粥样硬化斑块回声的超声分类的一个新特征。

Percentage white: a new feature for ultrasound classification of plaque echogenicity in carotid artery atherosclerosis.

机构信息

Wallenberg Laboratory for Cardiovascular Research, and Center for Cardiovascular and Metabolic Research, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden.

出版信息

Ultrasound Med Biol. 2010 Feb;36(2):218-26. doi: 10.1016/j.ultrasmedbio.2009.10.002. Epub 2009 Dec 16.

Abstract

Atherosclerotic stenotic and nonstenotic plaques of the carotid artery with low echogenicity have been shown to be associated with cardiovascular disease. The aim was to develop a new method for semiautomated ultrasound image analysis to classify nonstenotic carotid plaques, evaluate cases with multiple plaques and examine the association between a new image analysis feature of echogenicity and predictors of cardiovascular disease. The new image analysis feature, percentage white (PW), represents the fraction of bright structures inside a plaque and is integrated in an objective semiautomated method to evaluate echogenicity (SAMEE) in carotid plaques. PW was constructed to take into account overall echogenicity of the image as well as noise surrounding the plaque. Consecutive ultrasound examinations of carotid plaques from a population-based screening of 64-year-old women with varying risk for cardiovascular disease were selected for the present project; 92 far-wall and 47 near-wall plaques were used as a training dataset to develop the SAMEE algorithm with visual classification according to Gray-Weale as reference; 273 plaques were used to validate the method. All plaques were included in an analysis relating predictors of cardiovascular to average PW in all plaques, PW in the biggest plaque and to the plaque with lowest PW in each subject, respectively. In the training dataset the intermethodological variability between SAMEE and visual classification showed a kappa of 0.78 and a sensitivity and specificity of 96% and 81%, respectively. In the validation set, SAMEE and visual classification showed a kappa of 0.77, a sensitivity of 96% and a specificity of 80%. The reproducibility of PW was high, evidenced by r = 0.96 and CV = 9.85% at repeated examinations. Average PW values were associated with several predictors of cardiovascular risk: lipoprotein (a), HbA1c, blood glucose, apolipoproteinB/apolipoproteinA-I; and associated negatively with the levels of adiponectin and apolipoprotein A-I. In conclusion, PW integrated within a SAMEE is a new feature for assessment of echogenicity in carotid plaques and shows excellent reproducibility and agreement with visual assessment.

摘要

已经证明,颈动脉粥样硬化狭窄和非狭窄斑块的低回声与心血管疾病有关。目的是开发一种新的半自动化超声图像分析方法,对非狭窄性颈动脉斑块进行分类,评估多发性斑块病例,并研究新的回声分析特征与心血管疾病预测因素之间的关系。新的回声分析特征“白色百分比”(PW)代表斑块内亮结构的分数,并集成在一种客观的半自动化方法中,以评估颈动脉斑块的回声(SAMEE)。PW 的构建考虑了图像的整体回声以及斑块周围的噪声。本研究从一项针对 64 岁女性的基于人群的心血管疾病筛查中选择了连续的颈动脉斑块超声检查,这些女性的心血管疾病风险不同;92 个远壁斑块和 47 个近壁斑块被用作训练数据集,以根据 Gray-Weale 视觉分类开发 SAMEE 算法作为参考;273 个斑块用于验证该方法。所有斑块均纳入分析,将心血管疾病的预测因素与所有斑块的平均 PW、最大斑块的 PW 以及每个受试者中 PW 最低的斑块分别相关联。在训练数据集中,SAMEE 和视觉分类之间的方法间变异性的kappa 值为 0.78,敏感性和特异性分别为 96%和 81%。在验证集中,SAMEE 和视觉分类的 kappa 值为 0.77,敏感性为 96%,特异性为 80%。PW 的可重复性很高,在重复检查中 r = 0.96,CV = 9.85%。平均 PW 值与几种心血管风险预测因素相关:脂蛋白(a)、HbA1c、血糖、载脂蛋白 B/载脂蛋白 A-I;与脂联素和载脂蛋白 A-I 水平呈负相关。总之,集成在 SAMEE 中的 PW 是评估颈动脉斑块回声的一个新特征,具有出色的可重复性和与视觉评估的一致性。

文献AI研究员

20分钟写一篇综述,助力文献阅读效率提升50倍

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索