QUANTITATIVE TEXTURE ANALYSIS IN TWO-DIMENSIONAL ECHOCARDIOGRAPHY - APPLICATION TO THE DIAGNOSIS OF EXPERIMENTAL MYOCARDIAL CONTUSION
QUANTITATIVE TEXTURE ANALYSIS IN TWO-DIMENSIONAL ECHOCARDIOGRAPHY - APPLICATION TO THE DIAGNOSIS OF EXPERIMENTAL MYOCARDIAL CONTUSION
复制标题
DOI：
10.1161/01.cir.68.1.217
复制
发表时间：
1983-01-01
期刊：
影响因子：
37.8
通讯作者：
KERBER, RE
中科院分区：
文献类型：
--
作者：
SKORTON, DJ;COLLINS, SM;KERBER, RE
据推测，在2维（2D）超声心动图的区域图像纹理的分析将是一个准确的方法来区分正常和异常的心肌结构。这一假设进行了测试与定量纹理措施，以研究区域，空间分布的回波幅度在2D超声心动图进行之前和之后立即钝性左胸创伤诱导6麻醉犬。创伤后，心肌挫伤区出现更明亮，并表现出改变心肌纹理。通过使用一组定量纹理测量，当正常区域与挫伤区域相比时，在创伤前图像中没有发现显著差异。此外，在创伤前与创伤后图像中比较每只动物的正常区域时，没有发现差异。然而，12项测量区分了创伤后图像中的正常与挫伤区域（通过多变量方差分析，P值范围为0.0057-0.0001）。这些纹理措施能够区分正常与异常组织只有当纹理沿着方位角（横向）方向计算。显然，心肌挫伤区域表现出明显改变的局部回波幅度模式（改变的图像纹理），这些图像纹理改变可以用数字图像分析技术量化。定量纹理计算可能是一种有用的方法，超声组织定征。
It was postulated that the analysis of regional image texture in 2-dimensional (2D) echocardiograms would be an accurate method to differentiate normal from abnormal myocardial structure. This hypothesis was tested with quantitative texture measures to study the regional, spatial distribution of echo amplitudes in 2D echocardiograms performed before and immediately after blunt left chest trauma was induced in 6 anesthetized dogs. After trauma the contused region of myocardium appeared brighter and exhibited an altered myocardial texture. By use of a set of quantitative texture measures, no significant differences were found in pretrauma images when normal regions were compared with regions to be contused. Also, no difference was found when the normal regions in each animal were compared in pretrauma vs. posttrauma images. Twelve measures, however, differentiated normal from contused regions within the posttrauma images (P values ranged from 0.0057-0.0001 by multivariate analysis of variance). These texture measures were capable of differentiating normal from abnormal tissue only when texture along the azimuthal (lateral) direction was calculated. Apparently, regions of myocardial contusion exhibit visibly altered local echo-amplitude patterns (altered image texture), and these image texture alterations may be quantified with digital image analysis techniques. Quantitative texture calculations may be a useful approach to ultrasound tissue characterization.