Skip to main content

Volume 18 Supplement 1

19th Annual SCMR Scientific Sessions

Can we predict the diffusion "sweet-spot" based on a standard cine?


Cardiac diffusion tensor imaging (cDTI) is often performed using a stimulated echo sequence with monopolar diffusion encoding [1, 2]. While diffusion measured with this sequence is modified by strain [1] the standard strain model predicts that the strain effects can be eliminated by imaging at one of two "sweet-spots" in the cardiac cycle where the effects of strain cancel [3]. However, analysis of strain data to calculate the sweet-spot timings (SST) is typically too time consuming to be performed during an exam. Here we hypothesise that the SST is linearly related to the time to peak strain (PST) and PST will be approximately equal to end-systole time (EST) estimated from a standard cine. If so, EST could be used to easily predict SST.


2D spiral cine DENSE data [4] from a mid-ventricular short axis slice in 13 healthy subjects was analysed using the DENSE analysis tool from the University of Virginia [5]. Data were acquired on a Siemens Skyra with either a 3D (navigator gated, 0.10 cycles/mm balanced encoding, n = 5) or a 2D encoded protocol (breath hold, 0.06 cycles/m simple encoding), both 30 ms temporal/3.5 × 3.5 × 8 mm3 spatial resolution. Global radial and circumferential strain curves were analysed. The SST were defined as the time points during systolic contraction and diastolic relaxation where the strain was closest to the mean strain over the cardiac cycle. Radial and circumferential PST were averaged. EST was estimated as the timing of the image with minimum left ventricular blood pool from a 25 frame retrospectively gated short axis bSSFP cine in a similar plane to the DENSE images.


The figure (a and b) shows example radial and circumferential strain curves from one subject and the location of the SSTs. The systolic SST occurs later in the cardiac cycle when predicted using radial strain (p < 0.05) and the diastolic SST occurs earlier in the cardiac cycle using radial strain (p < 10-5). Diastolic and systolic SST calculated from either the radial or circumferential strain curves show a good correlation with the PST (figure c and table).

The correlation between SST and EST was poor (figure c and table), due to poor correlation between PST and EST (R2 = 0.007).


The diastolic and systolic SST can be predicted using the PST, but the locations vary depending on whether they are calculated using circumferential or radial strain. Radial strain is the larger component and has a greater influence on cDTI, but circumferential strain is more robust. The SST cannot be predicted from the EST, due to the unexpected result that EST does not relate to PST. Future work should consider whether online feature tracking could be used to predict SST. In the absence of strain information sweet-spot cDTI acquisitions should be performed at the mean time to sweet-spot.

Table 1 Results of the linear regressions of sweet-spot times with time to peak strain and end systolic time
Figure 1

Radial (a) and circumferential (b) strain-time curves from one example subject showing the location of the sweet-spots. Scatter plots of the sweet-spot time with time to peak strain (c) and the end-systolic time estimated from a short-axis cine (d) are shown for both the systolic and diastolic sweet spots calculated from both the radial and the circumferential strains.


  1. 1.

    Reese : MRM. 1995, 34: 786-

    Article  CAS  PubMed  Google Scholar 

  2. 2.

    Nielles-Vallespin : MRM. 2013, 70: 454-

    Article  PubMed  Google Scholar 

  3. 3.

    Tseng : MRM. 1999, 42: 393-

    Article  CAS  PubMed  Google Scholar 

  4. 4.

    Zhong : MRM. 2010, 64: 1089-

    PubMed Central  Article  PubMed  Google Scholar 

  5. 5.

    Spottiswoode : IEEE TMI. 2007, 26: 15-

    CAS  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Andrew D Scott.

Rights and permissions

This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated.

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Scott, A.D., Ferreira, P., Nielles-Vallespin, S. et al. Can we predict the diffusion "sweet-spot" based on a standard cine?. J Cardiovasc Magn Reson 18, W17 (2016).

Download citation


  • Diffusion Tensor Imaging
  • Cardiac Cycle
  • Circumferential Strain
  • Radial Strain
  • Feature Tracking