Hybrid Interleaved Multi-contrast Imaging (HIMI) for Simultaneous Brain and Carotid Vessel Wall Imaging
Shuo Chen1, Zechen Zhou1, Rui Li1, Xihai Zhao1, Huijun Chen1, Changwu Zhou1,2, Bida Zhang3, and Chun Yuan1,4

1Center for Biomedical Imaging Research, Department of Biomedical Engineering, School of Medicine, Tsinghua University, Beijing, China, People's Republic of, 2Department of Radiology, Yangzhou First People's Hospital, Yangzhou, China, People's Republic of, 3Healthcare Department, Philips Research China, Shanghai, China, People's Republic of, 4Vascular Imaging Laboratory, Department of Radiology, University of Washington, Seattle, WA, United States


The aim of this study was to develop a Hybrid Interleaved Multicontrast Imaging (HIMI) sequence for simultaneous brain and carotid vessel wall imaging. The proposed HIMI sequence takes advantage of the long delay time in conventional 3D FLAIR sequence to acquire multi-contrast carotid vessel wall images. Four healthy volunteers were recruited in this study. The results indicate that HIMI can generate a comparable FLAIR image with conventional FLAIR sequence and three more different contrast weighted (T1w, T2w, gray blood) carotid vessel wall images with the same scan time as a single conventional 3D FLAIR sequence.


Ischemic stroke is characterized by neurologic deficits that result from disruption of blood supply to the brain. Carotid atherosclerotic vulnerable plaque is a major cause of ischemic stroke [1]. Therefore, it is recommended that both the brain lesions and carotid atherosclerosis plaques should be evaluated for ischemic stroke patients [2]. While techniques as 3D FLAIR [3,4] and multi-contrast vessel wall imaging (VWI) sequences [5,6] demonstrate the benefits in brain imaging and plaque vulnerability evaluation, the low efficiency and long scan time limit the clinical application of these techniques. Notably, 3D FLAIR sequence contains a long delay time after the TSE echo train which is used for signal recovery. This long delay time has the potential to acquire additional carotid vessel wall images.


The aim of this work was to develop a 3D Hybrid Interleaved Multicontrast Imaging (HIMI) sequence which can provide (1) 3D FLAIR brain images, (2) intrinsically co-registered multi-contrast carotid vessel wall images, (3) acceptble SAR value (different location excited in interleaved order), and (4) high scan efficiency (<7min, same scan time with the single 3D FLAIR sequence).


Pulse sequence:

As shown in Figure 1, the proposed HIMI sequence consists of three acquisition modules which are performed in an interleaved order at different spatial locations to avoid local SAR problem. At the first module, 3D FLAIR brain image and T2w carotid VWI can be acquired using a coronally oriented TSE sequence which is similar to conventional 3D FLAIR sequence covering the whole brain and carotid artery. Saturation pulse (element 2) located around the carotid bifurcation and a DANTE pulse (element 4) are used to improve the SNR and blood suppression of T2w VWI. At the second module, T1w VWI can be acquired using a transversely oriented TFE sequence located around the bifurcation. Saturation pulse (element 6) and slab selective DANTE (element 7, Figure 2) are used to suppress the blood signal. Theoretically, intraplaque hemorrhage will show a hyperintense signal on T1w image. At the third module, a gray blood carotid image can be acquired using a TFE sequence without blood suppression prepulses. The gray blood image is potentially useful for calcification identification.

In vivo imaging:

To evaluate the feasibility of the proposed sequence, four healthy volunteers (male, mean age 24.8 years) were scanned on a 3.0T scanner (Achieva TX, Philips Medical System, Best, Netherlands) with a 16-channel neurovascular coil. Both proposed HIMI sequence and conventional 3D FLAIR sequence are scanned. The scan time for HIMI sequence and 3D FLAIR sequence is identical (6:46 mins). Imaging parameters are shown in Figure 3.

Data analysis:

Axial T2w carotid vessel wall images were reformatted with same resolution and location with T1w VWI and gray blood images from images acquired by HIMI module 1. For both HIMI and conventional 3D FLAIR, image qualities were scored by two reviewers using Likert scale (1-4 with 4 being the best). Gray matter – white matter (GM-WM) CNR and white matter – CSF (WM-CSF) CNR from HIMI and conventional 3D FLAIR were compared using a paired t-test.


All four volunteers were successfully scanned with proposed HIMI sequence and conventional 3D FLAIR sequence. Imaging time was 6:46 minutes for both methods. The HIMI sequence showed comparable FLAIR image and co-registered multi-contrast carotid vessel wall images (Figure 4). Good image quality were generated by HIMI: FLAIR (3.5±0.53 vs 3.875±0.35 conventional 3D FLAIR), T2w VWI (3.375±0.52), T1w VWI (3.25±0.46), gray blood (3.5±0.53) (Figure 5). Quantitative CNR measurement showed no significant difference in GM-WM CNR (5.56±0.96 vs 5.80±1.17, p=0.59) and GM-WM CNR (8.25±1.83 vs 8.9±2.05) between HIMI-FLAIR and conventional 3D FLAIR (Figure 5).

Discussion and conclusion

The study demonstrated that HIMI is a promising technique for simultaneous brain and carotid vessel wall imaging. With HIMI sequence, 3D FLAIR brain images and intrinsic co-registered 3D multi-contrast carotid vessel wall images can be acquired within 7 minutes. Except for brain infarction which can be detect by HIMI-FLAIR, carotid plaque components can potentially be determined by HIMI multi-contrast vessel wall images. To the best of our knowledge, the proposed HIMI sequence is the first technique that can simultaneous image intracranial lesions and responsible carotid plaques.


No acknowledgement found.


[1] Golledge J, et al. Stroke. 2000; 31:774–781.

[2] Latchaw RE, et al. Stroke. 2009; 40: 3646-3678.

[3] Kallmes DF, et al. Radiology. 2001;221:251–55.

[4] Chagla GH, et al. Invest Radiol. 2008;43:547–51.

[5] Saam T, et al. Arterioscler Thromb Vasc Biol. 2005; 25:234–39.

[6] Cai JM, et al. Circulation. 2002; 106:1368–73.


Figure 1. Diagram of the proposed HIMI sequence. Different spatial excitation regions of the indexed element within each sequence module are correspondingly shown. T2-FLAIR (module1) brain image and multi-contrast carotid vessel wall images (T2w with module1, T1w with module2, gray blood with module3) can be acquired using HIMI sequence.

Figure 2. Diagram of DANTE pulse (a) and proposed slab selective DANTE (ssDANTE) pulse (b).

Figure 3. Imaging parameters of 3D FLAIR sequence and HIMI sequence.

Figure 4. Typical healthy volunteer result: images (b, c, d, e) of HIMI sequence and conventional 3D FLAIR image (a). The HIMI-FLAIR shows a comparable FLAIR image with conventional 3D FLAIR sequence and three different contrast weighted carotid artery vessel wall images.

Figure 5. Qualitative and quantitative results of visual score (a) and tissues CNR (b,c). Image quality of HIMI sequence is good. No significant differences are found in GR-WM CNR and WM-CSF CNR between HIMI-FLAIR and conventional FLAIR.

Proc. Intl. Soc. Mag. Reson. Med. 24 (2016)