ausblenden:
Schlagwörter:
-
Zusammenfassung:
Purpose/Introduction: Experiments combining transcranial magnetic stimulation (TMS) and fMRI have recently received increased attention [1]. However, the standard set-up used for these measurements presents several limitations. To overcome them, we developed a thin 7-channel MR head coil array (see Fig. 1a) designed to be placed between the TMS system and the subject’s head [2]. Another critical issue for these experiments is the definition of a sham
condition [3]. This is required to have control over confounding TMS effects, like the ones produced on the auditory or somatosensory system. In this work, a viable set-up for a sham condition is proposed, and its influence on MR data quality is investigated.
Subjects and Methods: The sham condition was defined placing a 3 cm thick hollow block (an empty TMS stimulator housing) between TMS and MR coil (see Fig. 1b), which should prevent stimulation on the target while producing the same surrounding conditions such as vibrations or sounds. FLASH images of a spherical phantom were acquired at three different echo times (TE = 2.45/12.5/32.5 ms, TR = 400 ms, FoV = 181 9 223 mm2, resolution = 0.7 9 0.7 9 3 mm3, 10 slices, FA = 41, BW= 332 Hz/pixel) in the normal and the sham condition in a 3T MR Scanner (Tim Trio, Siemens, Erlangen, Germany). SNR maps were calculated for the three echo times and the two measurement conditions. SNR maps were co-registered to a coordinate system relative to the imaging coil, smoothed using a 3 mm Gaussian kernel, and masked to the phantom area. Relative SNR loss in the sham condition versus the normal condition was calculated.
Results: The increased distance to the TMS coil changes the tuning and matching conditions of the MR coil. Figure 2 and Table 1 show the relative SNR loss when using the sham condition. Results are shown for 3, 4.2 and 5 cm depth, and for the three echo times. A test participant reported subjectively similar noise and vibration levels for
real and sham stimulation.
Discussion/Discussion/Conclusion: The presented sham condition is suitable to be used in future interleaved TMS/fMRI experiments. For the typical application in fMRI, where echo times are around 30 ms at 3T, the mean SNR losses are lower than 10 . The gain in SNR for the sham
condition can be explained by decreased susceptibility effects caused by the TMS coil, when using the block between TMS and the MR head coil. SNR losses are attributed to MR coil detuning when removing the TMS. The MR head coil array could be tuned and matched to a more general condition in order to minimize them.
