Clinical Application of Simultaneous Multi-Slice DWI: A Multi-Body-Part Comparison with Standard EPI DWI
Man Sun1, Hongyan Ni2, Xu Yan3, and Tianyi Qian4

1First Central Clinical College, Tianjin Medical University, Tianjin, China, People's Republic of, 2Tianjin First Central Hospital, Tianjin, China, People's Republic of, 3MR Collaborations NE Asia, Siemens Healthcare, Shanghai, China, People's Republic of, 4MR Collaborations NE Asia, Siemens Healthcare, Beijing, China, People's Republic of


Applying simultaneous-multi-slice EPI (SMS EPI) sequences in diffusion-weighted imaging (DWI) can reduce scanning time greatly, but its clinical-routine image quality in different body parts has not been broadly tested. This study compared multi-site DWI images acquired with a prototype SMS EPI sequence with those acquired with a traditional EPI sequence, computed SNR and CNR and conducted paired t-test. The results show there are no significant differences of SNR and CNR between the two kinds of sequence (p>0.05). When the SMS EPI sequence was applied to the clinical DWI examination, it not only reduced scanning time, but also kept the SNR, CNR in the same level as standard EPI. The SMS sequence was able to satisfy the requirements of clinical disease diagnosis.


Slice-accelerated echo planer imaging (EPI) 1using special (multiband) RF pulses allows for simultaneous multi-slice (SMS) acquisition of MR signal. Compared with the traditional EPI (t-EPI) sequence, the SMS EPI sequence can greatly reduce repetition time (TR). If the SMS EPI sequence is applied in clinical routine examination, it not only reduces the waiting time for patients, but also improves examination efficiency, and decreases the possibility of motion artifacts during the examination. However, its clinical performance in different body parts has not been fully investigated. In this study, we applied SMS EPI on clinical multi-site (including head, abdominal, pelvis on female subjects, prostate on male subjects, and kidney) DWI to determine whether SMS EPI sequence performs well in several body parts.


SMS EPI prototype sequences and traditional EPI sequence data were acquired with routine head and body coils on a Siemens 3T MR Scanner (MAGNETOM Trio Tim, Siemens Healthcare, Erlangen, Germany). The participants were recruited from a clinical patient pool and all subjects gave written consent. The number of subjects was head (25), abdominal (13), pelvis on female (10), prostate on male (10) and kidney (10). For each body part, the parameters of the two types of DWI sequences are listed in table 1 (multi-slice factor =2). In order to have compatible image quality for the two sequences, we kept the same FOV, voxel size and TE and only reduced the TR in the SMS protocol. Signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) were calculated for the diffusion images collected by the two sequences. The region of interest (ROI) was manually drawn by two experienced radiologists. In order to test the repeatability, another retest measurement was taken two weeks after the first measurement. The results of the two time points were compared using Bland and Altman plot, and the results of the two sequence were compared by paired t-test.


The results of this study show that the diffusion image quality of the two kinds of sequences is comparable (fig.1). In DWI examination, although SNR of the SMS EPI sequence was a little lower than traditional EPI (t-EPI) sequence and CNR of SMS EPI sequence was higher than t-EPI sequence, the SNR and CNR differences between the two types of sequences showed no statistical significance (p>0.05) (take head for example see table 2). Meanwhile, the Bland and Altman plot shows that the repeatability of the two sequences is similar (take head for example see fig 2).


The main advantage of the simultaneous-multi-slice technique is decreased TR via multi-slice factor, and a shortened total acquisition time (see table.1). Decreasing acquisition time is highly beneficial for clinical MR examination, but it has to be confirmed that images obtained with SMS sequences will have the same diagnostic quality, SNR and CNR are key factors affecting image quality and disease diagnosis. In this study, the result show that, compared with the t-EPI sequence, the SNR and CNR of SMS EPI sequence do not have significant differences, and the repeatability of the two types of DWI sequence is very high.


Applying the SMS EPI sequence to the clinical DWI examination can greatly reduce scanning time and bring many benefits to patient and MR operator. The SNR and CNR do not have significant differences compared to standard EPI-based DWI. The SMS could replace the traditional EPI sequence in clinical routine work.


No acknowledgement found.


1. Feinberg, D. A, Moeller, S, Smith, S. M, et al. Multiplexed echo planar imaging for sub-second whole brain FMRI and fast diffusion imaging. PLoS One. 2010; 5(12):e15710


Table 1 SMS EPI sequence and traditional EPI sequence scan parameter for multi-site DWI examination

Fig.1: A-D is abdominal, pelvic, kidney and head DWI image. For every picture, the left to right is SMS EPI and traditional EPI sequence; For figure A, B, and C, from top to bottom is b=50, 800 diffusion image and ADC image; D is b=1000 diffusion image and ADC image.

Table 2 SNR and CNR of two types of head DWI sequence paired t-tests

Fig.2 The image SNR and CNR Bland-Altman plot for two measurements of head DWI image of two sequence: A and B, C and D show the SNR and CNR repeatability between two sequence; solid line is the mean difference for two measurements, dotted line is the 95% CI of difference.

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