This dataset contains Philips 3T PRESS data from 15 subjects, and can be found on NITRC.
I’ve added an entry for this dataset to the MRSHub data collection, and while I was there, created a new category for functional MRS data - I have a hunch that we’ll see more of these datasets coming out soon.
I’ve just read through the paper - it is stated there (and in the NITRC repository) that this is PRESS data at TE = 22 ms, which seems awfully short. Are you sure this isn’t STEAM data? Or maybe a typo in the TE?
No typo or STEAM, @erin.macmillan had to help me out with this one, but we managed to have an awfully short TE because we used a T/R head coil!
She explained to me that since this coil can transmit the RF pulses itself it can play them much faster, so the PRESS sequence is shorter leading to a shorter TE