Tissue Doppler Imaging (TDI) plays a major role in evaluating the left ventricular function [1]. This modality is characterized by an intrinsic noise, which leads to the classical thickness of the graphs displayed in commercial ultrasonic equipment. The agreement between TDI and Mitral Annulus Plane Excursion (MAPE) has been shown to be sub-optimal [2]. We hypothesize that this discordance is due to the noise in TDI. By taking the neighborhood space expectancy of the Tissue Doppler through time as the velocity of the Mitral Annulus Plane we obtain a periodic measurement which agrees better with the MAPE as measured manually. To validate our results we integrated the obtained velocity over time and obtained the cumulative displacement of the Mitral Annulus Plane. The Euclidean error between this displacement and manually marked MAPE was found to behave as a Gaussian distribution with mean 0 and standard deviation of 0.8 under QQ-plots and Kolmogorov-Smirnov test with significance levels of 5%.
GE Healthcare