Uicc fer PERF37 1330 TlilWeU Fui I Ouar ar OOOBlue

FIGURE 11 Regionalized strain measurements for 72 regions in a canine LV shown from base (top) to apex (bottom) and with anterior regions to the left and inferior regions to the right. Two strain measurements are plotted over the course of contraction, before and after partial occlusion of the left anterior descending (LAD) artery. Note the substantial differences between curves near the anterior apex. [Courtesy of E. McVeigh.] From Magnetic Resonance Imaging, 14, Elliot McVeigh, MRI of Myocardial Function: Motion Tracking Techniques, 137-150, © 1996 with permission from Elsevier Science. See also Plate 57.

advancements is needed. The first possibility is producing smarter and faster algorithms to perform the steps outlined in this chapter, which research has shown to produce valuable results. Toward this end, incremental steps are steadily being made and improved algorithms will continue to appear in the literature.

The second possible technical advancement is improvements in image acquisition that make high-resolution 3D imaging of LV contraction viable. Such 3D images would usher in a new approach to tagged MR image processing, eliminating the need for the motion estimation step discussed in Section 4. In particular, both HARP and optical flow techniques are inherently extendible to 3D image processing. At present, though, 3D image acquisition is prohibitively long for cardiac applications, and the approaches discussed within this chapter will remain the most likely solutions for detailed clinical assessment of myocardial motion.

0 0

Post a comment