2 次元超音波計測融合血流解析における解析精度の時空間変動特性

抄録

A two-dimensional ultrasonic-measurement-integrated (2D-UMI) blood flow analysis system was developed for easy acquisition of an intravascular hemodynamics, which feeds back Doppler velocity obtained by an ultrasonic measurement to a numerical blood flow simulation for clinical application. In previous study, ultrasonic measurement and 2D-UMI analysis were performed to clarify the analysis accuracy for real flow field, and the reproducibility of 2D-UMI analysis for real flow field was confirmed by comparison of velocity profile between 2D-UMI and 3D-CFD analysis results corresponding to an experimental flow. However, the analysis accuracy of 2D-UMI analysis for real flow has not been examined, and its spatiotemporal variation is important in evaluation of errors due to noise specific to ultrasonic measurement. The aim of this study was to clarify spatiotemporal variation characteristics of analysis accuracy of each velocity component obtained in 2D-UMI analysis. Comparisons of Doppler velocity V and (u, v) velocity profiles between measurement data, 2D-UMI, and 3D-CFD analyses results were performed, and their spatiotemporal variations were discussed. As a result, it was clarified that spatial variation of Doppler velocity error for frame-averaged measurement data became large when feedback gain is too large, and that of velocity vector error for 3D-CFD analysis result became smaller with increasing feedback gain. Temporal variations of Doppler velocity V, u, and v velocity profile became larger with increasing feedback gain, and it is considered that the analysis result for KV^* = 110 which the time variation is small and the velocity profile is smooth while reproducing the velocity profile of 3D-CFD analysis is close to an actual flow field.