Nelson T R, Verba J W, Bhargava V, Shabetai R, Slutsky R
Invest Radiol. 1983 Mar-Apr;18(2):130-7. doi: 10.1097/00004424-198303000-00004.
The purpose of this study was to analyze, validate, and report on an automatic computer algorithm for analyzing left ventricular ejection fraction and to indicate future applications of the technique to other chambers and more advanced measurements. Thirty-eight patients were studied in the cardiac catheterization laboratory by equilibrium radionuclide ventriculography and concurrent contrast ventriculography. The temporal and spatial behavior of each picture element in a filtered stroke volume image series was monitored throughout the cardiac cycle. Pixels that met specific phase, amplitude, and derivative criteria were assigned to the appropriate chamber. Volume curves were generated from regions of interest for each chamber to enable calculation of the left ventricular ejection fraction. Left ventricular ejection fractions showed a good correlation (r = 0.89) between the two techniques. Ejection fractions ranged between 0.12 and 0.88, showing a wide range of application. It is concluded that automatic analysis of left ventricular ejection fraction is possible using the present algorithm and will be useful in improving the reproducibility and providing more accurate information during exercise protocols, pharmaceutical interventions, and routine clinical studies.
本研究的目的是分析、验证并报告一种用于分析左心室射血分数的自动计算机算法，并指出该技术未来在其他心腔及更高级测量中的应用。在心脏导管实验室中，对38例患者进行了平衡放射性核素心室造影及同步对比心室造影研究。在整个心动周期中监测滤波后每搏量图像序列中每个像素的时空行为。符合特定相位、幅度和导数标准的像素被分配到相应的心腔。从每个心腔的感兴趣区域生成容积曲线，以计算左心室射血分数。两种技术测得的左心室射血分数显示出良好的相关性（r = 0.89）。射血分数范围在0.12至0.88之间，显示出广泛的应用范围。得出的结论是，使用当前算法可以自动分析左心室射血分数，这将有助于提高可重复性，并在运动试验、药物干预和常规临床研究中提供更准确的信息。
