Automated assessment of color Doppler M-mode early diastolic flow propagation features

摘要

Color Doppler M-mode echocardiographic imaging has been shown to provide useful information in the evaluation of left ventricular diastolic filling and function. However, descriptors of the spatiotemporal velocity inflow pattern do not take advantage of the full digital velocity map and are not available in an automated fashion. The purpose of this study was to automate the quantification of flow propagation features and evaluate their ability to assess diastolic filling abnormalities. Isovelocity contouring and gradient analysis were applied to identify features of the early filling wave. These feature parameters were compared in normal and diastolic dysfunction patient groups (N=20). Several features allowed classification of the two patient groups with sensitivity and specificity greater than 80%. This automated process may allow these features to be obtained on a regular basis and provide information for diagnosis and assessment of diastolic dysfunction.