One-minute whole-brain magnetization transfer ratio imaging with intrinsic B -correction.

PURPOSE

Magnetization transfer ratio (MTR) histograms are used widely for the assessment of diffuse pathological changes in the brain. For broad clinical application, MTR scans should not only be fast, but confounding factors should also be minimized for high reproducibility. To this end, a 1-minute whole-brain spiral MTR method with intrinsic B -field correction is introduced.

METHODS

A spiral multislice spoiled gradient-echo sequence with adaptable magnetization-transfer saturation pulses (angle β) is proposed. After a low-resolution single-shot spiral readout and a dummy preparation period, high-resolution images are acquired using an interleaved spiral readout. For whole-brain MTR imaging, 50 interleaved slices with three different magnetization-transfer contrasts (β = 0°, 350°, and 550°) together with an intrinsic B -field map are recorded in 58.5 seconds on a clinical 3T system. From the three contrasts, two sets of MTR images are derived and used for subsequent B correction, assuming a linear dependency on β. For validation, a binary spin bath model is used.

RESULTS

For the proposed B -correction scheme, numerical simulations indicate for brain tissue a decrease of about a factor of 10 for the B -related bias on MTR. As a result, following B correction, MTR differences in gray and white matter become markedly accentuated, and the reproducibility of MTR histograms from scan-rescan experiments is improved. Furthermore, B -corrected MTR histograms show a lower variability for age-matched normal-appearing brain tissue.

CONCLUSION

From its speed and offering intrinsic B correction, the proposed method shows excellent prospects for clinical studies that explore magnetization-transfer effects based on MTR histogram analysis.