Bing Wu1, Jianbo Shao2, Xuehua Peng2, Bing Wu3, Mingmei Ge3, Xinhuai Wu3, Hui Lin1, and Zhenyu Zhou1
1GE healthcare MR Research China, Beijing, People's Republic of China, 2Wuhan Children Hospital, 3PLA Army General Hospital
Synopsis
In this study, the use use of reduced FOV CUBE
is studied in comparison to full FOV CUBE as well as other common techniques.
It was seen that CUBE sequences provide robust and superior image details and
black blood effects, due to the advantageous 3D FSE acquisition. With only 1/4
of the FOV acquired, reduced FOV CUBE may complete the scan using half of the
scan time (using twice Nex) while obtaining highly consistent results as FF
CUBE. Reduced FOV CUBE may also reduce the motion related artifacts that may
affect the homogeneity of the lumen area.
Purpose
Intracranial plaque is a leading cause of hemorrhage, the
screening and diagnosis of plaque need both morphological and compositional
assessment. The former requires the use of high resolution black blood acquisition
to differentiate between plaque, vessel wall and lumen, the latter requires multi-contrast
acquisitions for differentiation among hemorrhage, lipid core, calcification
and fiber[1]. 3D FSE based acquisition, such as SPACE and CUBE, has
been shown to be suitable for plaque imaging due to their high SNR, low
distortion and black blood effects. Furthermore, it is capable of T1, T2 and T1
enhanced acquisitions that facilitate image registrations. However, conventional
3D FSE acquisition is challenged by the relatively long acquisition time, which
not only reduces the patient comfort level and also increases the motion vulnerability.
Since plaque imaging is in general only concerned with the bifurcation of the
arteries, the actual volume of interest is rather limited. In this work, we
investigate the use of reduced field of view (rFOV) 3D FSE acquisition [2]
for intracranial plaque imaging.Method
4 patients pre-diagnosed with plaque using ultrasound or CT
angiography participated in this study, consent forms were obtained. All the
participants underwent the following scans on a 3.0T scanner (MR 750, GE) using
an 8-channel dedicated plaque coil: MSDE, QIR FSE, MPRAGE, full FOV (FF) CUBE T1,
reduced FOV (RF) CUBE T1, FF CUBE T2, RRF CUBE T2 (Table 1). Image analysis was focused on two aspects: the contrast
to noise ratio (CNR) and dimensions of the plaque. The CNRs were calculated as
(plaque-lumen)/lumen for all the sequences; the dimensions of the plaque were
measured slice-wise and compared between CUBE and rFOV CUBE. Lumen and plaque
regions were identified semi-automatically using a signal intensity based
regional growth approach by an experienced radiologist. The ROIs were
identified on CUBE and copied to different contrasts, given identical FOV and
in plane resolutions were used.Results
An axial slice with the presence of plaque in the same
patient from different sequences are shown in Fig.1. It is seen that CUBE acquisitions provided relatively higher
level of SNR; CUBE acquisitions also led to superior suppression of the blood
as seen by dark lumen areas. The measured CNRs for MSDE, QIR, MPRAGE, FF CUBE
T1 and RF CUBE T1 were 21.3, 23.5, 22.7, 33.4, 36.2 respectively. The sagittal
slice images of the same patient of full FOV and reduced FOV CUBE T1, T2, T1
contrast are shown in Fig.2. It can
be seen that morphological dimensions as well as different compositions were
well depicted, the hypo-intense region in contrast enhanced T1 indicated the
presence of calcification. Fig.3
illustrates a case where motion induced artifacts were present in the lumen
(circled), which may be mistaken as plaque. The arrowed region also suggested
much stronger motion artifacts in the FF CUBE. The scatter plot of the measured
lumen areas in different axial slices obtained using full FOV and reduced FOV
acquisitions is shown in Fig.4, a
high correlation coefficient of 0.94 was obtained.
Discussion and conclusion
Intracranial plaque imaging conveys great clinical significance,
and many techniques have been developed for its needs. In this study, the use use
of reduced FOV CUBE is studied in comparison to full FOV CUBE as well as other
common techniques. It was seen that CUBE sequences provide robust and superior image
details and black blood effects, due to the advantageous 3D FSE acquisition. With
only 1/4 of the FOV acquired, RF CUBE may complete the scan using half of the
scan time (using twice Nex) while obtaining highly consistent results as FF
CUBE. In this study, even higher lumen CNR was obtained using RF CUBE, as the
lumen areas in FF CUBE may be contaminated by motion artifacts in FF CUBE. The
decreased scan time may be used improving patient comfort and reducing
vulnerability to motion as demonstrated, or higher spatial resolution may be
obtained improving the depiction of the vascular structures. The multi-contrast
compatibility of RF CUBE and its short scan time (that facilitates image
registration) would be of great clinical value of intracranial plaque imaging.Acknowledgements
No acknowledgement found.References
[1] J.
Wang, et al., Simultaneous
Noncontrast Angiography and intraPlaque
hemorrhage
(SNAP) Imaging for Carotid Atherosclerotic Disease Evaluation, MRM, 2013:69.
[2] M. Han, et al., Variable flip angle
three-dimensional fast spin-echo sequence combined with outer volume
suppression for imaging trabecular bone structure of the proximal femur. JMRI,
2015:41(5).