Background suppressed magnetization transfer MRI

• Published in: Magnetic resonance in medicine Vol. 83; no. 3; pp. 883 - 891
• Main Authors: Gelderen, Peter, Duyn, Jeff H.
• Format: Journal Article
• Language: English
• Published: United States Wiley Subscription Services, Inc 01.03.2020
• Subjects: Adult; Algorithms; background suppression; Brain - diagnostic imaging; Brain Mapping; exchange; Exchanging; Humans; Hydrogen; Hydrogen atoms; Image Processing, Computer-Assisted - methods; Imaging, Three-Dimensional; Magnetic Resonance Imaging; Magnetics; Magnetization; magnetization transfer; Motion; myelin; Semisolids; Signal processing; Signal-To-Noise Ratio; stimulated echo; Subtraction; Thermal noise; Time dependence; Water; white matter; Young Adult; myelin; white matter; magnetization transfer; stimulated echo; exchange; background suppression
• ISSN: 0740-3194; 1522-2594; 1522-2594
• DOI: 10.1002/mrm.27978

Purpose Up to 30% of the hydrogen atoms in brain tissue are part of molecules (“semisolids”) other than water. In MRI, their magnetization is typically not observed directly, but can influence the water magnetization through magnetization transfer (MT). Comparison of MRI scans differentially sensitized to MT allows estimation of the semisolid fraction and potential changes with disease. Here, we present an approach designed to improve this estimate by measuring the size of the MT effect in a single scan. Methods A stimulated echo sequence was used to generate a spatial pattern in the longitudinal water magnetization, which was then given time to exchange with semisolids. After saturating the remaining water magnetization, reverse exchange was allowed to partly re‐establish the original water magnetization pattern. The third excitation pulse then formed a stimulated echo out of this pattern. Results MT data were obtained on 10 human subjects at 7 T with varying exchange times. The images showed the expected time dependence of signal associated with the forward and reverse exchange processes. Excellent suppression of non‐exchanging background signal was achieved. As expected, this suppression came at the price of a substantial reduction in exchange‐related signal (by ~75% compared to the signal in saturation recovery MT), in part because of the reliance on a 2‐step exchange process. Conclusion The results demonstrate an MT signal can be observed in a single acquisition without subtraction. This may be advantageous for MT measurements when signal instabilities related to motion and physiological variations exceed thermal noise sources.