Gaillard E, Kadem L, Pibarot P, Durand L-G
Laboratory of Biomedical Engineering, Montreal Clinical Research Institute (IRCM), University of Montreal, Montreal, Canada.
Annu Int Conf IEEE Eng Med Biol Soc. 2009;2009:2264-7. doi: 10.1109/IEMBS.2009.5334968.
Intra- and inter-observer variability in Doppler velocity echocardiographic measurements (DVEM) is a significant issue. Indeed, imprecisions of DVEM can lead to diagnostic errors, particularly in the quantification of the severity of heart valve dysfunction. To minimize the variability and rapidity of DVEM, we have developed an automatic method of Doppler velocity wave contour detection, based on active contour models. To validate our new method, results obtained with this method were compared to those obtained manually by an experienced echocardiographer on Doppler echocardiographic images of left ventricular outflow tract and transvalvular flow velocity signals recorded in 30 patients, 15 with aortic stenosis and 15 with mitral stenosis. We focused on three essential variables that are measured routinely by Doppler echocardiography in the clinical setting: the maximum velocity, the mean velocity and the velocity-time integral. Comparison between the two methods has shown a very good agreement (linear correlation coefficient R(2) = 0.99 between the automatically and the manually extracted variables). Moreover, the computation time was really short, about 5s. This new method applied to DVEM could, therefore, provide a useful tool to eliminate the intra- and inter-observer variabilities associated with DVEM and thereby to improve the diagnosis of cardiovascular disease. This automatic method could also allow the echocardiographer to realize these measurements within a much shorter period of time compared to standard manual tracing method. From a practical point of view, the model developed can be easily implanted in a standard echocardiographic system.
多普勒速度超声心动图测量（DVEM）中观察者内和观察者间的变异性是一个重要问题。事实上，DVEM的不精确性可能导致诊断错误，尤其是在心脏瓣膜功能障碍严重程度的量化方面。为了最小化DVEM的变异性和提高其速度，我们基于主动轮廓模型开发了一种自动多普勒速度波轮廓检测方法。为了验证我们的新方法，将该方法获得的结果与一位经验丰富的超声心动图医生在30例患者（15例主动脉瓣狭窄和15例二尖瓣狭窄）的左心室流出道多普勒超声心动图图像和跨瓣膜流速信号上手动获得的结果进行了比较。我们重点关注了临床环境中通过多普勒超声心动图常规测量的三个重要变量：最大速度、平均速度和速度时间积分。两种方法之间的比较显示出非常好的一致性（自动提取变量与手动提取变量之间的线性相关系数R² = 0.99）。此外，计算时间非常短，约为5秒。因此，这种应用于DVEM的新方法可以提供一个有用的工具，以消除与DVEM相关的观察者内和观察者间的变异性，从而改善心血管疾病的诊断。与标准手动追踪方法相比，这种自动方法还可以使超声心动图医生在更短的时间内完成这些测量。从实际角度来看，所开发的模型可以很容易地植入标准超声心动图系统中。
