Chemical shift encoding using asymmetric readout waveforms. Issue 3 (8th October 2020)

• Record Type: Journal Article
• Authors: Rydén, Henric; Norbeck, Ola; Avventi, Enrico; Skorpil, Mikael; van Niekerk, Adam; Skare, Stefan; Berglund, Johan
• Is Part Of: Magnetic resonance in medicine. Volume 85:Issue 3(2021)
• Page Start: 1468
• Page End: 1480
• Publication Date: 2020-10-08
• Subjects: chemical shift -- Dixon -- MRI; Nuclear magnetic resonance -- Periodicals; Electron paramagnetic resonance -- Periodicals; 616.07548
• DOI: 10.1002/mrm.28529
• Languages: English
• ISSNs: 0740-3194

Abstract

Abstract : Purpose: To describe a new method for encoding chemical shift using asymmetric readout waveforms that enables more SNR‐efficient fat/water imaging. Methods: Chemical shift was encoded using asymmetric readout waveforms, rather than conventional shifted trapezoid readouts. Two asymmetric waveforms are described: a triangle and a spline. The concept was applied to a fat/water separated RARE sequence to increase sampling efficiency. The benefits were investigated through comparisons to shifted trapezoid readouts. Using asymmetric readout waveforms, the scan time was either shortened or maintained to increase SNR. A matched in‐phase waveform is also described that aims to improve the SNR transfer function of the fat and water estimates. The sequence was demonstrated for cervical spine, musculoskeletal (MSK), and optic nerve applications at 3T and compared with conventional shifted readouts. Results: By removing sequence dead times, scan times were shortened by 30% with maintained SNR. The shorter echo spacing also reduced T 2 blurring. Maintaining the scan times and using asymmetric readout waveforms achieved an SNR improvement in agreement with the prolonged sampling duration. Conclusions: Asymmetric readout waveforms offer an additional degree of freedom in pulse sequence designs where chemical shift encoding is desired. This can be used to significantly shorten scan times or to increase SNR with maintained scan time.