i-fMRI: A virtual whiteboard discussion on multi-echo, simultaneous multi-slice EPI

Disclaimer: This isn't an April Fool!

I'd like to use the collective wisdom of the Internet to discuss the pros and cons of a general approach to simultaneous multislice (SMS) EPI that I've been thinking about recently, before anyone wastes time doing any actual programming or data acquisition.

Multi-echo EPI for de-noising fMRI data

There has been quite a lot of interest in using multi-echo EPI to characterize and de-noise time series data, e.g.

http://www.ncbi.nlm.nih.gov/pubmed/22209809

These methods rest on one critical aspect: they use in-plane parallel imaging (GRAPPA or SENSE, usually depending on the scanner vendor) to render the per slice acquisition time reasonable. For example, with R=2 acceleration it's possible to get three echo planar images per slice at TEs of around 15, 40 and 60 ms. The multiple echoes can then be used to characterize BOLD from non-BOLD signal variations, etc.

The immediate problem with this scheme is that the per slice acquisition time is still a lot longer than for normal EPI, meaning less brain coverage. The suggestion has been to use MB/SMS to regain speed in the slice dimension. This results in the combination of MB/SMS in the slice dimension and GRAPPA/SENSE in-plane, thereby complicating the reconstruction, possibly (probably) amplifying artifacts, enhancing motion sensitivity, etc. If we could eliminate the in-plane parallel imaging and do all the acceleration through MB/SMS then that would possibly reduce some of the artifact amplification, might simplify (slightly) the necessary reference data, etc.

A different approach? 

What I've been mulling is using the PRESTO approach of echo shifting in combination with MB/SMS. Right now we assume that we are crushing all residual signal at the end of each slice acquisition whatever EPI scheme we're using. But if instead we set up the gradients at the end of each readout period so that they are fully balanced/refocused prior to the next excitation then we might have some signal from the prior slice(s) existing at the same time as new signal from the subsequent slice(s). In an experiment detected with a non-array RF coil this would just cause artifacts, but with an array RF coil and the MB/SMS scheme it seems like a solvable problem. Essentially, at the time of acquisition there would be the primary signals from the current SMS excitation, all arising from one set of slice positions, plus the secondary (old) signals from the prior SMS excitation arising from a different set of slice positions. For simplicity, if we assume just these two types of signal are present during each readout period then a modified MB/SMS reconstruction should be able to separate the signals. And we would get two different effective TEs to use in characterizing the BOLD-like characteristics.

Caveats

I can think of two immediate problems with this idea. The first problem is that the extended TE - that for the residual signal from the prior excitation - will be more than twice the prescribed TE, making the recycled signal quite weak. We might expect to use a primary TE of around 30 ms, say, but the secondary TE would then be around 80 ms once we've accounted for fat saturation and the other housekeeping activities. (We could probably set up the gradient balancing at the end of a slice readout quite efficiently such that little or no additional time is needed to preserve the signal for the subsequent readout.) Is such dual echo data even useful for characterizing BOLD from non-BOLD? (A dual TE experiment on a limited number of slices, no SMS, would be a way to answer that question.)

The second problem is that some signals are likely to persevere for three or more acquisition periods; CSF and vitreous fluid come to mind because they have very long T2 and quite long T2*. Will the persistence of these uninteresting signals overwhelm the potential benefits of second echo signals from gray matter? We might also get accidental spin and stimulated echoes arising from any imperfections in the slice selection scheme, too.


Over to you

So there you have today's thought experiment. If nothing else perhaps this is a good way for you to learn more about EPI - accelerated or not - as it is acquired today. It might make an interesting teaching tool for understanding how gradients balance and how k-space works. Can you draw out the modified gradient scheme that would be required to maintain signal from excitation to excitation in a single-shot unaccelerated EPI sequence?



A note on nomenclature: Simultaneous multi-slice (SMS) and multi-band (MB) are used equivalently here. However, I am trying to shift towards using SMS rather than MB because I think it is a more intuitive description of the sequence, and it's hard enough to comprehend the alphabet soup in MRI as it is!