Hi,
We have collected 1H MRS data in motor cortex with semi-Laser sequence with TE=68ms and 64 averages on a Siemens 7T scanner, after the automated prepocess pipeline with Osprey (including ECC) , we found the NAA peak is still asymmertric. I wonder if we need do some manual correction to make the processed data looks better?
Here is the processed spectra:
Best,
Chuyue
I agree that the dip at 1.9 ppm looks a little off, but the rest of the spectrum looks almost perfectly phased. I would expect that the modeling will work just fine here.
Do you have a basis set? I don’t think we ship 7T sLASER for that TE with Osprey.
Hi,
Thank you for the reply. I think I don’t have a porper basis set now. I tried to use basisset produced by MRScloud, but I found it is simulated as B0=3T. So I’m trying to simulate basis set with FID-A, but I am now confused about which function should I use to simulate basis set for semi-Laser seuqnece.
Best,
Chuyue
Hi Chuyue,
Your non-averaged data looks very noise at around 1.9ppm. But because of the little gap I would rather suspect that one half of your FIDs is lower and the other one higher. The lower part probably causes this asymmetry. Maybe you have to make a plot with your FIDs with different colors…
Best,
Heiner
Hi Heiner,
Thank you for the reply, I ploted non-averaged raw spectra and it looks like there are some differents patterns of spectra between 1.5 ppm to 2.0 ppm. But I don’t know how to interprete this differents and how to correct this.
Here I plot every 4 single spectra in a sub figure:
And here is differents patterns of spectra between 1.5 ppm to 2.0 ppm:
Best,
Chuyue
I’m not too familiar with the Siemens 7T platform, but a few thoughts:
Hi alex,
Thank you for the advice, I tried skip ECC but the sharp dip around 2.65 ppm is still here with the odd shape around 1.9 ppm
Hi Chuyue,
I also wondered about the sharp dip that Alex mentioned but I’m not familiar with 7T. I also suggest that both artifacts are related. For a rolling baseline from phase effects other parts of the spectrum look too good. But I wouldn’t rule it out, as the spectrum between 1 and 2ppm also looks a bit strange. Spectra can also be distorted when there is an ADC overflow (see spectrum 10, NMR Artifacts).
Just to be sure, is this systematic across multiple subjects, or is it just one peculiar dataset? If it’s just one dataset, I would be wary of selective extra, manual (non-repeatable) processing…
I’d be curious to see what happens when you try to fit it.
We scanned two different subject, both of the subjects have 2 MRS scan with TE=42ms and TE=68 ms respectively. But all the processed spectra showed sharp dip around 2.65 ppm and 1.9 ppm.
This might just be J modulation of the aspartyl signal. I dug up this figure of an [old paper of mine]( Low visual cortex GABA levels in hepatic encephalopathy: links to blood ammonia, critical flicker frequency, and brain osmolytes | SpringerLink) where we analyzed 68-ms PRESS data. I don’t find it entirely unreasonable to assume that something like that happens here, but obviously we’d need to simulate.
@NSniper that brings me to the point - it will not be super straight forward to make that basis set for you. I think this is the CMRR sequence, and I do not have the sequence timings or pulse shapes for that sequence. Someone in your institution must have a so-called C2P agreement with the Minnesota people - I think the best thing would be for this person in your institution to reach out to them and ask for a matching basis set. (If all else fails, I can probably e-mail them on your behalf)
Thank you for the advice, I will try to check the artifact problem.
Thank you for the reply, I think Ican read the paper but I’m not very clear about how to simulate this effect. We used the CMRR sequence and have a C2P agreement. I’ll ask them for the matching basis set.
