This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. The subproject and investigator (PI) may have received primary funding from another NIH source, and thus could be represented in other CRISP entries. The institution listed is for the Center, which is not necessarily the institution for the investigator. Introduction: While magnetization transfer (MT) is useful in detecting subtle changes in white matter diseases, its use at high field strengths has been limited by SAR concerns [1]. Most MT sequences use SAR-intensive RF pulses, or use complex parametric models requiring large datasets from RF-spoiled gradient-echo (GRE) acquisitions. White matter pool (WIMP) mapping uses a stimulated echo (STE) method by Ropele et al. [2] combined with a variable density (VD) spiral readout [3] that directly measures the bound proton fraction (BPF). However, WIMP uses the difference between two similar-contrast STE images, which results in low signal (and thus SNR) in the computed BPF images. In experiments at 1.5T, an average of only <3% difference between the images was noted. While several averages are simple solution to increase SNR, with excessive averaging motion becomes an increasing issue. Both additional signal and signal difference may be possible via the use of higher field strength, e.g. 3T. Ropele s method is particularly interesting for high field applications since, aside from exceptionally low SAR, it requires neither the use of MT RF pulses nor the knowledge of underlying the T1 or T2 values. This method is therefore both feasible and practical at high field strength. Methods: A WIMP sequence is similar to a stimulated echo preparation, with three ?/2 RF pulses. The second and third pulses are separated by a time TM, which includes a composite refocusing pulse of two abutted 2.4 ms ?/2 pulses. The phase of the second pulse can be modulated, resulting in a total composite flip angle of 0 or 180 . A chemically-selective RF pulse and gradient spoiling immediately precede the final RF pulse to perform fat saturation, as in Fig. 1. VD spiral readouts then follow this BPF-mapping WIMP preparation. Acquisitions in a volunteer were performed on a GE Signa 3.0T scanner (GEHC, Waukesha, WI) with 40mT/m gradients using the standard quadrature birdcage head coil, and an 8 channel phased array head coil (MRI Devices, Milwaukee, WI). Scan parameters are: FOV =24cm, matrix=256x256, TR=3s, TE=6ms, TM=200ms, NEX=10, thickness=5mm for both examinations. The WIMP labeling gradients were 15 mT/m and lasted 250 ?sec. The variable density spiral used 16 interleaves with pitch factor 2.5. In order to calculate the relative bound pool size (SWIMP) and the bound pool fraction (BPFWIMP), the sequence is run twice. BPF images attained with the quadrature head coil are presented in the top row of Fig. 2; while those acquired with the receive-only phased-array coil are in the bottom row. As the quadrature head coil is used for B1 transmission, rather than the body coil for the receive-only coil, its smaller diameter, even at 3T, gives a more marked B1 variation across the volume of interest. The residual B1 artifacts are evident in the final quantitative BPF maps on the top right in Fig. 2. The more even excitation profile of the body coil aids the image quality in the bottom right of Fig. 2. Despite the image apodization from the B1 field, image quality is remarkable, showing excellent grey/white contrast. BPF values in white matter are consistent with those previously reported [4-6].
Showing the most recent 10 out of 446 publications
