Although the diagnostic and scientific value of these motion functions is well established, in vivo determination of LV motion from medical images is a major challenge. The primary problem is that the myocardium is quite uniform, providing few readily identifiable features in images. As demonstrated in Fig. 1, the inner (endocardial) and outer (epicardial) boundaries of the LV are apparent, but little transmural variation is present. In such featureless regions, motion measurements are impossible. For the LV in particular, most conventional images only allow measurement of wall thickening [13].

The lack-of-features problem is effectively solved by a technique unique to MR imaging called tagging [1,2]. Tagging uses pulse sequences — similar to standard imaging where an are coefficients determined by the tip angles of the RF pulses. The magnitude of g determines the spatial frequency of the tag pattern and the direction of g determines the tag direction. The pattern in Fig. 2 is an example of SPAMM tags, where the dark lines appear wherever the cosines line up in phase. Finally, two SPAMM sequences are often applied in rapid succession with orthogonal gradient directions resulting in a grid pattern given by f N -1

where g 1 and g2 are the two gradient directions.

0 0

Post a comment