Multislice helical computed tomography (CT) substantially reduces scanning
time. However, the temporal resolution of individual images is still insuff
icient for imaging rapidly moving organs such as the heart and adjacent pul
monary vessels. It may, in some cases, be worse than with current single-sl
ice helical CT. The purpose of this study is to describe a novel image reco
nstruction algorithm to improve temporal resolution in multislice helical C
T, and to evaluate its performance against existing algorithms. The propose
d image reconstruction algorithm uses helical interpolation followed by dat
a weighting based on the acquisition time. The temporal resolution, the lon
gitudinal (z-axis) spatial resolution, the image noise, and the in-plane im
age artifacts created by a moving phantom were compared with those from the
basic multislice helical reconstruction (helical filter interpolation, HFI
) algorithm and the basic single-slice helical reconstruction algorithm (18
0 degrees linear interpolation, 180LI) using computer simulations. Computer
simulation results were verified with CT examinations of the heart and lun
g vasculature using a 0.5 second multislice scanner. The temporal resolutio
n of HFI algorithm varies from 0.28 and 0.86 s, depending on helical pitch.
The proposed method improves the resolution to a constant value of 0.29 s,
independent of pitch, allowing moving objects to be imaged with reduced bl
urring or motion artifacts. The spatial (z) resolution was slightly worse t
han with the HFI algorithm; the image noise was worse than with the HFI alg
orithm but was comparable to axial (step-and-shoot) CT. The proposed method
provided sharp images of the moving objects, portraying the anatomy accura
tely. The proposed algorithm for multislice helical CT allowed us to obtain
CT images with high temporal resolution. It may improve the image quality
of clinical cardiac, lung, and vascular CT imaging. (C) 2000 American Assoc
iation of Physicists in Medicine.