Measurement of in vivo multi-component T2 relaxation times for brain tissue using multi-slice T2 prep at 1.5 and 3 T

• Published in: Magnetic resonance imaging Vol. 24; no. 1; pp. 33 - 43
• Main Authors: Oh, Joonmi, Han, Eric T., Pelletier, Daniel, Nelson, Sarah J.
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier Inc 2006
• Subjects: Adult; Algorithms; Brain Mapping - methods; Female; High-field MR; Humans; Least-Squares Analysis; Magnetic Resonance Imaging - instrumentation; Magnetic Resonance Imaging - methods; Male; Multi-component T2 relaxation times; Multi-slice multi-echo images; Phantoms, Imaging; Quantification; T2 prep; Multi-slice multi-echo images; High-field MR; Multi-component T2 relaxation times; T2 prep; Quantification
• ISSN: 0730-725X; 1873-5894
• DOI: 10.1016/j.mri.2005.10.016

The objective of this study was to implement a clinically relevant multi-slice multi-echo imaging sequence in order to quantify multi-component T2 relaxation times for normal volunteers at both 1.5 and 3 T. Multi-echo data were fitted using a nonnegative least square algorithm. Twelve echo data with nonlinear echo sampling were acquired using a receive-only eight-channel phased array coil and volume head coil for phantoms and normal volunteers, and compared to 32-echo data with linear echo sampling. It was observed that the performance of the 180° refocusing trains was more spatially uniform for the receive-only eight-channel phased array coil than for the head coil, particularly at 3 T. The phantom study showed that the estimated T2 relaxation times were accurate and reproducible for both single- and multi-slice acquisition from a commercial phantom with known T2 relaxation times. Short T2 components (T2 <50 ms) were mainly observed within the white matter for normal volunteers, and the fraction of short T2 water components (i.e., myelin water) was 7–12% of total water. It was observed that the calculated myelin water fraction map from the nonlinearly sampled 12-echo data was comparable with that from the linearly sampled 32-echo data. Quantification of T2 relaxation times from multi-slice images was accomplished with a clinically acceptable scan times (16 min) for normal volunteers by using a nonselective T2 prep imaging sequence. The use of the eight-channel head coil involved more accurate quantification of T2 relaxation times particularly when the number of echoes was limited.