Source localization with LCMV AND dSPM

[georgekenefati]

Hi all,

Is there a way to create a pipeline that creates filters using LCMV, then computes the source time courses using dSPM?

Two papers for reference:

• Linear constraint minimum variance beamformer functional magnetic resonance inverse imaging
• Comparison of beamformer implementations for MEG source localization

Thanks!
George

[jstout211 (Jeff Stout)]

@georgekenefati

filters using LCMV, then computes the source time courses using dSPM?

I don’t think works. They are both methods to map the data to the brain. So you can’t map it to the brain and then map it from the brain to the brain again.

You can use t/SSS which is a spatial filter that results in channel data and then map the channel data to the brain with either of those methods.

Those papers generally contrast LCMV and dDSPM as separate methods.

Hope that helps.
–Jeff

[jstout211 (Jeff Stout)]

I just thought about this a little more. I guess you could project hte data down to the brain with LCMV, then project that out to the sensors and then map it back to the brain with dSPM. But there may be some underlying parameters that may be unhappy. So it likely wouldn’t be turnkey and you likely wouldn’t benefit much from it.