hello,
Do any variables in MRSContainer of Osprey track the amplitude of the water peak in the signal? In the QM table, I see a 'residual_water_amplitude" - is this something I could use as an approximate of the Water signal in my MRS voxel? I use HERCULES and my data are water suppressed but I was hoping changes in residual water amplitude over time might be a proxy for the Water signal.
The reason I want to do this is: I am analyzing my data in moving window of 96 transients and hoping to check how the water / ‘bold’ signal changes over time during my acquisition. I have simultaneous EEG and know that some subjects are falling asleep during the acquisition and expect to see a difference in the oxygen signal in my voxel during the sleep scans.
I also saw a post on this forum that VDI software can extract a BOLD/Water signal from an MRS voxel but I am having trouble finding an example script that shows which functions to use to do this.
I don’t have a complete answer, but a few initial thoughts:
I believe Osprey typically normalises the incoming data, so depending on how the residual_water_amplitude is scaled, it’s possible that the normalisation may cancel any of the real amplitude changes you expect, and residual_water_amplitude simply becomes an indication of the water suppression efficiency. I’m not certain, but you should check this.
Residual water can be extremely sensitive to subject motion, which may introduce a substantial confound, especially if some of the subjects are drifting off to sleep mid-scan…
Some of the HERCULES editing pulses are fairly close to water, which may exacerbate these issues. You probably need to be careful about the choice of subspectra for measuring residual water.
If residual water amplitude doesn’t work out for you, note that many studies look at change in linewidth (of unsuppressed water if you have it, or major metabolite peaks if not) as an indication of BOLD-related T2* effects. (refer Zhu & Chen 2001)