Electron Beam CT Versus 16-MDCT on the Variability of Repeated Coronary Artery Calcium Measurements in a Variable Heart Rate Phantom
Jun Horiguchi1,
Yun Shen2,
Yuji Akiyama1,
Nobuhiko Hirai1,
Kousuke Sasaki2,
Minoru Ishifuro1,
Tadashi Nakanishi3 and
Katsuhide Ito4
1 Department of Clinical Radiology, Hiroshima University Hospital, 1-2-3,
Kasumi-cho, Minami-ku, Hiroshima 734-8551, Japan.
2 Imaging Application Tech Center, GE Yokogawa Medical Systems, Tokyo,
Japan.
3 Department of Radiology, Mazda Hospital, Mazda Motor Corporation, Hiroshima,
Japan.
4 Department of Radiology, Division of Medical Intelligence and Informatics,
Programs for Applied Biomedicine, Graduate School of Biomedical Sciences,
Hiroshima University, Hiroshima, Japan.
View larger version (77K):
[in a new window]
|
Fig. 1 —Photograph (left) shows cardiac phantom, and graphic (right)
shows balloon. Phantom consisted of five components: driver, control, support,
rubber balloon, and ECG. A controller with an ECG-synchronizer drove the
balloon. The motion was achieved by setting four driver sequences—that
is, two speeds of fast emptying for the systolic phase and fast and slow
filling for the diastolic phase. The balloon was filled with a mixture of
water and contrast medium (58 H) to simulate noncontrast blood and was
submerged in corn oil (-118 H), simulating epicardial and pericardial fat.
Coronary artery calcium models were packed inside rubber tubes (mimicking
coronary arteries) attached to the balloon surface. The ends of the balloon
were stabilized to a fixed support at a distance of 10 cm. There was therefore
neither through-plane motion (along z-axis) nor twist motion of the
balloon. The volumes of the balloon phantom were approximately 100 and 200 mL
at the systolic and diastolic phases, respectively. The time-balloon volume
curve was similar to sinusoidal in heart rate shift sequences. The balloon was
barrel-shaped at the diastolic phase and nearly, but not exactly, cylindrical
at the systolic phase. Deformity of the balloon was seen in some images in
high heart rate and arrhythmia sequences. This movement of the balloon
resulted in some through-plane motion of calcium models.
|
|
View larger version (94K):
[in a new window]
|
Fig. 3 —Graph shows medians (z) and means (bars) of
variability of Agatston score. Vertical lines show range: mean + SD
(top) and mean -SD (bottom). Helical CT shows lower
variability than axial CT; 0.625-mm helical CT showed lowest variability. Use
of thin-slice images and overlapping reconstruction improves reproducibility.
EBCT = electron beam CT.
|
|
View larger version (104K):
[in a new window]
|
Fig. 4 —Graph shows mean variability of Agatston score in three heart
rate sequences: 60 beats per minute (bpm) (white bars), 85 bpm
(gray bars), and arrhythmia (black bars). No difference is
seen between heart rates of 60 and 85 bpm on either electron beam CT or
16-MDCT. EBCT = electron beam CT.
|
|
View larger version (113K):
[in a new window]
|
Fig. 5 —CAC measurement algorithm. Graph shows mean variability by
Agatston score (white bars), volume score (gray bars), and
mass score (black bars). Reduction of variability by volume scoring
algorithm was effective only on electron beam CT (p = 0.05). Mass
scoring algorithm was effective on electron beam CT (p < 0.01),
2.5-mm helical CT (p < 0.01), and 2.5-mm axial CT (p =
0.02). Variability in volume and mass measurements for EBCT was almost the
same as for helical CT protocols.
|
|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Copyright © 2005 by the American Roentgen Ray Society.
), and arrhythmia (
).