Search

Objective: To compare three-dimensional (3D) and two-dimensional (2D) speckle tracking echocardiography (STE) techniques in the assessment of left ventricular function and myocardial infarct size (MIS)., Methods: Thirty-two patients diagnosed with ST elevation myocardial infarction and 18 healthy control patients underwent 2D echocardiography, 3D echocardiography, and single photon emission computed tomography (SPECT). 3D left ventricular global area strain (GAS), 2D and 3D global longitudinal strain (GLS), global radial strain (GRS) as well as global circumferential strain (GCS) were analyzed to correlate with myocardial infarct size detected by SPECT. 2D and 3D left ventricular ejection fraction (LVEF) as well as 2D and 3D wall motion score index (WMSI) also were measured using conventional echocardiography., Results: The 2D-GLS values were significantly higher than that of 3D-GLS, while 2D-GCS and GRS were significantly lower than 3D-GCS and GRS, respectively. However, no significant differences in LVEF and WMSI could be observed between 2D and 3D echocardiography. Myocardial strain indices, LVEF, and WMSI using 2D and 3D echocardiography also had good correlations with MIS as measured by SPECT. ROC curve analysis showed that the 3D and 2D myocardial indices, LVEF, and WMSI could distinguish between small and large MIS, while 2D-GLS had the highest AUC., Conclusion: The 2D and 3D myocardial strain indices correlated well with MIS by SPECT. Among them, the 2D-GLS showed the highest diagnostic value, while 3D-GRS and GCS had better diagnostic value than 2D-GRS and GCS., (© 2015, Wiley Periodicals, Inc.)

Echocardiography is the first-line diagnostic technique for heart diseases. Although artificial intelligence techniques have made great improvements in the analysis of echocardiography, the major limitations remain to be the built neural networks are normally adapted to a few diseases and specific equipment. Here, we present an end-to-end deep learning framework named AIEchoDx that differentiates four common cardiovascular diseases (Atrial Septal Defect, Dilated Cardiomyopathy, Hypertrophic Cardiomyopathy, prior Myocardial Infarction) from normal subjects with performance comparable to that of consensus of three senior cardiologists in AUCs (99.50% vs 99.26%, 98.75% vs 92.75%, 99.57% vs 97.21%, 98.52% vs 84.20%, and 98.70% vs 89.41%), respectively. Meanwhile, AIEchoDx accurately recognizes critical lesion regions of interest along with each disease by visualizing the decision-making process. Furthermore, our analysis indicates that heterogeneous diseases, like dilated cardiomyopathy, could be classified into two phenogroups with distinct clinical characteristics. Finally, AIEchoDx performs efficiently as an anomaly detection tool when applying handheld device-produced videos. Together, AIEchoDx provides a potential diagnostic assistant tool in either cart-based echocardiography equipment or handheld echocardiography device for primary and point-of-care medical personnel with high diagnostic performance, and the application of lesion region identification and heterogeneous disease phenogrouping, which may broaden the application of artificial intelligence in echocardiography. [ABSTRACT FROM AUTHOR]