Abstract-When recording velocities using ultrasound Doppler, the obtained velocities are always a projection of the true velocities onto the direction from the recording point towards the transducer. In this study we have marked up lines in sequences of ultrasound images of the heart showing the direction of the original velocity. Doing this, it is possible to calculate the original velocity in the marked direction. Both the recorded and corrected velocities have been used for calculating strain, in order to compare the two. Apical ultrasound images have been recorded on a sample of 18 normal individuals. The corrected velocities are on the average less than 5% higher than the recorded ones, and the effect on the derived variable strain is even less, so we suggest that the correction can be omitted when recording tissue velocities through an apical window. Keywords -Echocardiography, Doppler, angular error, angle dependency, strain, strain rate, tissue velocity1 I. INTRODUCTION When recording velocities using ultrasound Doppler techniques, the velocity obtained is a projection of the true velocity vector on the line from the recording point to the transducer [1]. This causes an error, which increases with increasing angle for the direction of the true velocity. This error in turn affects other derived variables, such as strain rates [2] and strain.To make it possible to correct this, one must know the original direction of the velocity. When seeing the heart as a dynamic displacement pump [3], one can recognize the main direction of the myocardial motion as being longitudinal. When marking up the longitudinal direction in the ultrasound picture, it is possible to calculate the original velocities from the recorded ones.Both two-and four-chamber apical images have been recorded on a sample of 18 normal individuals. The study of tissue velocities in the myocard using an apical window may seem narrow, but this is in fact frequently used for the investigation of myocardial performance [4][5][6], as well as derived variables calculated using data acquired this way [7]. Thus both corrected and non-corrected velocities have been used for calculating strain, as peak systolic strain is an easily calculated variable. This has been taken as an example to show how the angular correction can affect derived variables.The work is financed by the Swedish Foundation for Strategic Research and the Swedish Heart Lung Foundation.
II. METHODOLOGY A. Raw dataThe raw data consisted of both two-and four-chamber sequences of ultrasound images recorded through an apical window on 18 normal individuals. A GE Vingmed System V machine has been used. Each sequence is recorded over two consecutive heartbeats acquired at rest at end expiratory apnea. An averaging over the two heartbeats has been done.
B. Extracting velocities and directionsVelocities have been extracted at six points in three pairs in each heart wall, enumerated 1-6 starting from apex towards the base as shown in fig. 1. The walls are the septal and lateral walls displayed in ...