Echo Planar Time-resolved Imaging (EPTI)
Fuyixue Wang1,2, Zijing Dong3, Timothy G. Reese1, Berkin Bilgic1, Mary Kate Manhard1, Lawrence L. Wald1,2, and Kawin Setsompop1,2

1A. A. Martinos Center for Biomedical Imaging, Department of Radiology, Massachusetts General Hospital, Charlestown, MA, United States, 2Harvard-MIT Health Sciences and Technology, MIT, Cambridge, MA, United States, 3Center for Biomedical Imaging Research, Department of Biomedical Engineering, Tsinghua University, Beijing, China


A new technique, termed Echo Planar Time-resolved Imaging (EPTI), was developed to address EPI’s geometric distortion and blurring, and to provide new temporal signal evolution information across the EPI readout window. Using only a few shots, a time-series of multi-contrast images can be created free of distortion and blurring (up to 100 T2- and T2*-weighted images with time interval of an EPI echo spacing). This should make EPTI useful for numerous applications where undistorted images across multiple-contrasts are desired. We demonstrated EPTI in brain to simultaneously map T2, T2*, and tissue phase, as well as to provide SWI.


Echo planar imaging (EPI) is a commonly-used MR acquisition technique due to its fast speed. However, there are two major problems with EPI: i) geometric distortion along the phase-encoding direction, and ii) its ability to only obtain a single-contrast image at the effective echo time, with blurring effects from other time points in the EPI readout. These problems significantly compromise EPI’s image quality in functional/diffusion/perfusion imaging, and limit its ability to achieve high-quality anatomical and quantitative imaging.

In this study, a new technique based on EPI, termed Echo Planar Time-resolved Imaging (EPTI), was developed to address these issues. This approach not only achieves distortion- and blurring- free imaging, it is also capable of obtaining up to 100 T2&T2*-weighted images across the EPI readout window, spaced at a time interval equal to EPI’s echo-spacing (~1ms). This should make it useful to numerous applications where high-SNR undistorted images or multiple-contrast images are desired. Here, we validated EPTI by applying it to simultaneously map T2, T2*, and tissue phase, as well as to provide susceptibility weighted imaging (SWI) in the brain.


To understand how EPTI works, a different perspective of EPI signal space, ky-t space, is introduced in Fig.1. For single-shot EPI (SS-EPI), the signal is acquired to fill a 45° diagonal line in the ky-t space, with T2/T2* decay and susceptibility-induced phase accumulating over time, leading to blurring and distortion in the final image. To correct for distortion, a pair of datasets with reversed phase-encoding is usually acquired[1,2]. Such acquisition obtains two +/-45° diagonal lines in the ky-t space, with more information to estimate and correct for the susceptibility-induced distortion (Fig.1a). To obtain multiple-contrast images, multi-echo EPI methods[3,4] can be used as shown in Fig1.b, but suffer from limited number of echoes as well as image distortions and blurring.

If we acquire data to fully-sampled ky-t space, distortion- and blurring- free images with different contrasts can be obtained at different echo times with a spacing-interval of an echo-spacing. Such fully-sampled ky-t data can be achieved through several existing techniques, such as EPSI[5] which acquires horizontal lines in ky-t space for different shots, and PSF-encoded EPI[6,7] which acquires multiple diagonal lines for different PSF encodings. However, these techniques require extremely long scans, especially for high-spatial resolution.

To achieve an optimal acquisition to resolve the temporal evolution of EPI signals, EPTI is proposed with “tilted-CAIPI” reconstruction[8]. As shown in Fig1.c, each EPTI-shot covers a segment of ky-t using a zig-zag trajectory with an interleaved acceleration in the phase-encoding direction. The zig-zag trajectory ensures that neighboring ky-points are acquired only a few milliseconds apart, and contain small B0-inhomogeneity induced phase and T2* decay that can be estimated well by parallel imaging and B0-inhomogeneity-informed reconstruction[8]. Here, the reconstruction utilizes compact kernels to interpolate under-sampled ky-t space to fully-sampled ky-t, which requires an acquisition of low-resolution calibration scan. Partial Fourier in ky can also be implemented into EPTI to reduce the number of shots/segments.

EPTI can be used for quantitative T2&T2* mapping, using single spin-echo or dual-echo (gradient-echo & spin-echo) acquisition (Fig.1c), and for SWI using the gradient-echo portion(s) of the same acquisition. Phantom and in-vivo experiments were performed at 3T to validate EPTI (see acquisition parameters in figure captions).


In phantom (Fig.2), EPTI achieves comparable T2 mapping to multi-TE acquisitions of standard-SE sequence, and more convincing T2* fittings compared to 8-echo GRE. Accelerated EPTI (8-shot, 27×) performed comparably to fully-sampled 216-shot acquisition, with slightly increased noise. In Fig.3A, multiple-contrast brain images from 8-shot EPTI are of similar quality to ones from fully-sampled data. In Fig.3B, combined GE and SE EPTI images show significant SNR-enhancement without distortion and blurring, when compared to SS-EPI. In Fig.4, simultaneous T2&T2* mapping in brain using EPTI shows close estimation to gold-standard SE and GRE mapping, at dramatically reduced acquisition time (20s across multiple slices compared to 28+5min). High quality tissue-phase and SWI can be obtained together with T2&T2* maps as per Fig.5.

Discussion and Conclusion

EPTI is an efficient technique for distortion- and blurring- free anatomical imaging, multiple-contrast imaging and quantitative imaging. In this study, EPTI allows simultaneous quantitative T2&T2* mapping and SWI in 30s at 1×1×3mm3 across multiple slices. This could translate to whole-brain coverage in ~30s if combined with SMS=2. EPTI can also be further combined with multi-inversion[9] for simultaneous T1, T2 and T2* mapping. Moreover, the multi-echo property of EPTI renders it applicable to many applications, such as for improving phase/SWI/QSM and fMRI through better signal modeling[10-13].


This work was supported in part by NIH research grants: R01EB020613, R01EB019437, R24MH106096, P41EB015896, and the shared instrumentation grants: S10RR023401, S10RR019307, S10RR019254, S10RR023043.


1. Zahneisen B, Aksoy M, Maclaren J, Wuerslin C, Bammer R. Extended hybrid-space SENSE for EPI: Off-resonance and eddy current corrected joint interleaved blip-up/down reconstruction. NeuroImage 2017;153:97-108.

2. Andersson JLR, Skare S, Ashburner J. How to correct susceptibility distortions in spin-echo echo-planar images: application to diffusion tensor imaging. NeuroImage 2003;20:870-888.

3. Kundu P, Inati SJ, Evans JW, Luh WM, Bandettini PA. Differentiating BOLD and non-BOLD signals in fMRI time series using multi-echo EPI. NeuroImage 2012;60:1759-1770.

4. Barth M, Reichenbach JR, Venkatesan R, Moser E, Haacke EM. High-resolution, multiple gradient-echo functional MRI at 1.5 T. Magn. Reson. Med. 1999;17:321-329.

5. Mansfield P. Spatial mapping of the chemical shift in NMR. Magn. Reson. Med. 1984;1:370-386.

6. Zeng H, Constable RT. Image distortion correction in EPI: comparison of field mapping with point spread function mapping. Magn. Reson. Med. 2002;48:137-146.

7. Robson MD, Gore JC, Constable RT. Measurement of the point spread function in MRI using constant time imaging. Magn. Reson. Med. 1997;38:733-740.

8. Dong Z, Wang F, Reese TG, Manhard MK, Bilgic B, Wald LL, Guo H, Setsompop K. Tilted-CAIPI for Highly Accelerated Distortion-Free EPI with Point Spread Function (PSF) encoding. Submitted to: In Proceedings of the 26th Annual Meeting of ISMRM 2018.

9. Cohen O, Polimeni JR. Optimized inversion-time schedules for quantitative T1 measurements based on high-resolution multi-inversion EPI. Magn. Reson. Med. 2017.

10. Wharton S, Bowtell R. Fiber orientation-dependent white mattercontrast in gradient echo MRI. Proc Natl Acad Sci USA 2012;109:18559–18564.

11. Cronin MJ, Wang N, Decker KS, Wei H, Zhu WZ, Liu C. Exploring the origins of echo-time-dependent quantitative susceptibility map-ping (QSM) measurements in healthy tissue and cerebral micro-bleeds. NeuroImage 2017;149:98–113.

12. Wang Y, Liu T. Quantitative susceptibility mapping (QSM): decoding MRI data for a tissue magnetic biomarker. Magn. Reson. Med. 2015;73: 82-101.

13. Kundu P, Voon V, Balchandani P, et al. Multi-echo fMRI: A review of applications in fMRI denoising and analysis of BOLD signals. NeuroImage, 2017.


Figure 1. The trajectories in ky-t space of ‘top-up’ distortion correction (a), multi-echo EPI (b), and EPTI (c). Each shot of EPTI is acquired in a zig-zag pattern, covering a segment of ky-t with small temporal distance (Nt) between adjacent phase encodings. The phase encoding and the number of shots are accelerated by factors of RPE and Rseg, respectively, as indicated in (c). EPTI is able to generate multiple distortion-free images with different contrasts with a time interval of an echo spacing (~1ms). T2 and T2* can be obtained by acquiring a single spin-echo or dual-echo as illustrated in (c).

Figure 2. (A) T2 maps and comparison of T2 values between spin-echo, fully-sampled ky-t data and 8-shot EPTI. The T2 ROI analysis validates the ability of EPTI to provide accurate T2 maps. (B) T2* maps and the corresponding fitting curves at voxel ‘i’ indicated by yellow arrows. Compared with 8-echo GRE, EPTI achieves a more convincing T2* fitting. The phantom was acquired at 1×1×3mm3 resolution, using the following parameters: 1) Standard SE: TEs=25,50,75,100,125,150,200ms; 2) multi-echo 2D GRE: TEs =10,20,30,40,50,60,70,80ms; 3) EPTI: number of shots=8, Partial Fourier=0.73, Rseg=20, RPE=4, Nt=5, number of echoes (GE/SE)=61/45, echo spacing=0.95ms, echo time range of GE/SE=17.3-74.3ms/84.3-126.1ms.

Figure 3. (A) Multi-contrast images (4 out of 94) reconstructed from fully-sampled ky-t data and 8-shot EPTI data (27×) are of similar quality. (B) Comparison of distorted SS-EPI and distortion-free EPTI SE and GRE images. Combined GE and SE EPTI images show significant SNR enhancement without distortion and blurring compared to SS-EPI. Data were acquired and reconstructed with the following parameters: number of shots=8, Partial Fourier=0.73, Rseg=20, RPE=4, Nt=5, number of echoes (GE/SE)=49/45, echo spacing=1.12/0.95ms, TR=2500ms, echo time range of GE/SE=19.6-73.4ms/84.3-126.1ms. Combined images were obtained by using sum of square of multi-echo images.

Figure 4. (a) T2 and (b) T2* maps from standard spin-echo and multi-echo GRE; (c) Simultaneous T2&T2* maps from EPTI using the same dataset as in Fig.3. Imaging parameters: 1) Standard SE: TR=1680ms, TEs=25,50,75,100,125,150,200ms; 2) multi-echo 2D GRE: TR=1050ms, TEs=10,20,30,40,50,60,70,80ms; 8 slices were acquired for validation. Acquisition efficiency was dramatically improved by EPTI, which obtains T2&T2* maps within 20s (for up to 18slices) compared to conventional SE (28min, 8slices) and GRE (5min, 8slices). Note, additional ~20s was needed for EPTI calibration, which should decrease through further optimization.

Figure 5. The combined magnitude, tissue phase and SWI of fully-sampled data and EPTI data, obtained using the gradient echo portion (19.6-73.4ms) of the same datasets as in Fig.3, but without partial Fourier (i.e. with additional 3 shots). High quality tissue phase and SWI can be acquired simultaneously together with T2&T2* maps by EPTI.

Proc. Intl. Soc. Mag. Reson. Med. 26 (2018)