Functional Magnetic Resonance Imaging in a Low-Field Intraoperative Scanner

Abstract

Background: Functional magnetic resonance imaging (fMRI) has been used for preoperative planning and intraoperative surgical navigation. However, most experience to date has been with preoperative images acquired on high-field echoplanar MRI units. We explored the feasibility of acquiring fMRI of the motor cortex with a dedicated low-field intraoperative MRI (iMRI). Methods: Five healthy volunteers were scanned with the 0.12-tesla PoleStar N-10 iMRI (Odin Medical Technologies, Israel). A finger-tapping motor paradigm was performed with sequential scans, acquired alternately at rest and during activity. In addition, scans were obtained during breath holding alternating with normal breathing. The same paradigms were repeated using a 3-tesla MRI (Siemens Corp., Allandale, N.J., USA). Statistical analysis was performed offline using cross-correlation and cluster techniques. Data were resampled using the ‘jackknife’ process. The location, number of activated voxels and degrees of statistical significance between the two scanners were compared. Results: With both the 0.12- and 3-tesla imagers, motor cortex activation was seen in all subjects to a significance of p < 0.02 or greater. No clustered pixels were seen outside the sensorimotor cortex. The resampled correlation coefficients were normally distributed, with a mean of 0.56 for both the 0.12- and 3-tesla scanners (standard deviations 0.11 and 0.08, respectively). The breath holding paradigm confirmed that the expected diffuse activation was seen on 0.12- and 3-tesla scans. Conclusions: Accurate fMRI with a low-field iMRI is feasible. Such data could be acquired immediately before or even during surgery. This would increase the utility of iMRI and allow for updated intraoperative functional imaging, free of the limitations of brain shift.

The Early Effect of Alcohol and Caffeine on a BOLD Signal Measured in Human Hand at Low-Field MRI

“…A good concordance and significance of the data was revealed, but many repetitions for data collection were required. Recently, a dedicated low-field intra-operative 0.12 T scanner was tested with a motor task and a hypoxia paradigm [14]. In that study, a 3-T MR scanner was used for reference measurement, again with good correlation, and the motor fMRI was proved to be possible at the low-field strength.…”

“…Thus, the paradigm used for the induced BOLD hand experiments would be easily transferable to brain fMRI in the future. Just recently, low-field MR systems have been shown to be capable of measuring fMRI for intraoperative purposes, with good correlation with the results from higher field strength studies, but significantly lower signal level [13,14]. Using alcohol or caffeine to emphasize the BOLD signal measured at low-field strengths might be very useful, especially if the initial effect of these substances is sufficient and no premeasurement delay is needed.…”

TTC post-processing is beneficial for functional MRI at low magnetic field: a comparative study at 1 T and 3 T

“…Despite some successful demonstrations of feasibility of intraoperative fMRI [23][24][25], appropriate fMRI sequences (fluid-attenuated inversion recovery [26] or gradient echo) and mostly post-processing strategies at low magnetic field require more investigation to cope with inherently low signal-tonoise ratio.…”

Advanced Intraoperative MR Imaging

“…Two healthy subjects were examined using an EPI sequence (TR / TE = 3000 / 80 ms) with 10 mm thick slices, while performing an index finger to thumb tapping task in a block design. In 2003, Schulder et al published a series of 5 subjects undergoing intraoperative fMRI using a very low field system [39]. Two years later intraoperative fMRI was performed using a 1.5 Tesla scanner and a T2*-weigthed EPI sequence (TR / TE = 3300 / 60 ms, slice thickness 3 mm) under general anesthesia with electrical stimulation of the median and tibial nerves [40].…”