Cg. Diederichs et al., BLURRING OF VESSELS IN SPIRAL CT ANGIOGRAPHY - EFFECTS OF COLLIMATIONWIDTH, PITCH, VIEWING PLANE, AND WINDOWING IN MAXIMUM INTENSITY PROJECTION, Journal of computer assisted tomography, 20(6), 1996, pp. 965-974
Purpose: Our goal was to examine the effects of collimation width (CW)
, pitch, viewing plane, and windowing on the display of in-plane vesse
ls in maximum intensity projection (MIP). Method: A theoretical concep
t based on partial volume averaging of vessels was developed to descri
be the contents of voxels (densities) in MIP and to derive cross-secti
onal vessel diameters and blurring. To validate the concept and to des
cribe the influence of pitch, a Plexiglas cone submerged in water was
scanned with varying CW and pitch. Binary MIP with three representativ
e window levels was chosen so that definitive vessel diameters could b
e quantitated. Results: The theoretical concept correctly predicted vo
xel contents and blurring for CW greater than or equal to 3 mm and low
pitch. For high pitch, actual blurring was larger; however, for a giv
en table speed, blurring of the cone decreased with pitch while increa
sing with CW. Overall blurring was most effectively reduced by using a
thin CW and the transverse viewing plane. In the transverse viewing p
lane, the least blurring was found using binary MIP with a low window
level. On the contrary, in the longitudinal viewing plane, blurring wa
s minimized using a window level halfway between the density of the co
ne and that of the surrounding water. Conclusion: For CW greater than
or equal to 3 mm, blurring of in-plane vessels can be explained with a
simple geometrical concept based on partial volume. For accurate disp
lay, the transverse viewing plane should be used, a proper windowing m
ust be chosen, and the CW should be kept below vessel size while raisi
ng the pitch to cover a reasonable volume.