Department of Biomedical Engineering, University of Alberta, Edmonton, Alberta, Canada.
Magn Reson Med. 2010 Oct;64(4):1005-14. doi: 10.1002/mrm.22487.
Presented is a fitting model for transverse relaxometry data acquired with the multiple-refocused spin-echo sequence. The proposed model, requiring no additional data input or pulse sequence modifications, compensates for imperfections in the transmit field and radiofrequency (RF) profiles. Exploiting oscillatory echo behavior to estimate alternate coherence pathways, the model compensates for prolonged signal decay from stimulated echo pathways yielding precise monoexponential T(2) quantification. Verified numerically and experimentally at 4.7 T in phantoms and the human brain, over 95% accuracy is readily attainable in realistic imaging situations without sacrificing multislice capabilities or requiring composite or adiabatic RF pulses. The proposed model allows T(2) quantitation in heterogeneous transmit fields and permits thin refocusing widths for efficient multislice imaging.
本文提出了一种适用于多回波自旋回波序列采集的横向弛豫率数据的拟合模型。该模型不需要额外的数据输入或脉冲序列修改，即可补偿发射场和射频（RF）轮廓的不完美。通过利用回波的振荡行为来估计替代相干途径，该模型补偿了来自受激回波途径的信号衰减，从而实现了精确的单指数 T(2)定量。在 4.7T 的体模和人脑上进行了数值和实验验证，在不牺牲多层面成像能力或不需要组合或绝热 RF 脉冲的情况下，该模型在实际成像情况下可实现超过 95%的准确度。所提出的模型允许在不均匀的发射场中进行 T(2)定量，并允许使用薄的重聚焦宽度来实现高效的多层面成像。
