Basis Set simulation: shaped refocus, ideal excitation... why?

Analysis / Modeling Basis Sets
1

Hi,

Many readily-available (sample) implementations for basis simulation with shaped refocussing pulses still use ideal excitation pulses (eg, MRSCloud, FID-A, Vespa). Why is this?

By my understanding,

  1. In many common cases the shaped excitation pulse has comparatively minor effects on the simulation outcomes
  2. …therefore, the additional complexity and processing time for simulating real excitation pulses was simply not considered worthwhile

This made sense when an extra spatial dimension implied an additional level of nested loops in an already lengthy simulation, but with the advent of density matrix simulation methods (thanks Zhang et al, 2017) this is no longer a real concern – the excitation dimension becomes just another 1D loop in the series, adding a very tolerable 50% to the already quite reasonable processing time… 10s of seconds at worst. Shaped excitation pulses are very much a part of Zhang et al’s method.

So my question is: why aren’t these more routinely used? The underlying tools are capable; are there some other technical/physical complexities which I’m not seeing, or is it more just a historic compromise (which it’s perhaps time to reconsider)?

1 Like
2

Hi Alex,

You’re right, there’s no real reason other than convenience, access to information, and indeed the higher bandwidths of the 90 pulses which reduces the spatial effects we seek to address. One additional non-trivial bit of info that you need is the ‘phase center’ of the excitation pulse because you need to include the rewinding gradient. Easy for symmetrical pulses, but a little less trivial for non-symmetrical pulses (like the Philips default 90).

Cheers,
Georg

4 Likes