Effect of Long TE onT1Measurement in STEAM Progressive Saturation Experiment
✍ Scribed by Jack Knight-Scott; Shi-Jiang Li
- Publisher
- Elsevier Science
- Year
- 1997
- Tongue
- English
- Weight
- 105 KB
- Volume
- 126
- Category
- Article
- ISSN
- 1090-7807
No coin nor oath required. For personal study only.
✦ Synopsis
Progressive saturation (1) is the most widely used longitu-sive saturation experiment. Results from simulations and measurements performed on a phantom are also presented. dinal relaxation measurement technique in localized clinical proton magnetic resonance spectroscopy. When flip angles
For simplicity, we will exclude coupling effects, and assume off-resonance and diffusion effects are negligible, all flip angles and frequency settings are near ideal, peak signal intensities from a single-pulse progressive saturation (PS) experiment are perfect 90Њ, spin-spin relaxation (T 2 ) dominates the natural transverse relaxation (T 2 ), i.e., T 2 É T 2 , and B 0 inhomogeneity can be shown to obey (1) is dominated by the application of crusher gradients. The influ-S (TD) Å S 0 [1 0 exp(0TD/T 1 )],
[1] ence of slice-selective gradients can be ignored if the gradients are balanced with rephasers to correct for dephasing caused by where S(TD) is the signal intensity as a function of the the application of these gradient pulses. Since we are interested delay time (TD), T 1 is the longitudinal relaxation time, and in relaxation effects, it is easier to follow the macroscopic mag-S 0 is the signal intensity when TD ӷ T 1 . For the original netization, i.e., the ''ensemble average,'' at particular points in single-pulse PS sequence, TD is equal to the repetition time time than to analyze the effects of the sequence on individual (TR), and Eq. [1] may be rewritten as spin systems where we must be concerned with coupling effects. Let M 0 represent the macroscopic thermal equilibrium mag-S (TR) Å S 0 [1 0 exp(0TR/T 1 )].
[2] netization, and M z and M xy the net macroscopic longitudinal and transverse magnetization in the rotating frame, respectively, for Recent 1 H MRS studies have measured the T 1 of metabolites a group of equivalent spins with I Å 1 2 . For the STEAM sequence using long echo times in the spectroscopic stimulated echo shown in Fig. 1, all pulses are 90Њ 0x . Immediately following acquisition mode (STEAM) localization sequence (2-7), and the first pulse at time t Å 0 / , the longitudinal magnetization is fitted to Eq. [2]. rotated into the transverse plane such that the longitudinal (z) Unfortunately, for multipulse sequences, Eq. [2] no longer and transverse (xy) components are given by accurately describes the longitudinal relaxation behavior, even M z (0 / ) Å 0 [3] if ideal conditions of perfect RF pulses and negligible off-resonance effects are assumed. For STEAM at short echoes times M xy (0 / ) Å M z (0 0 ). (TE) and short mixing times (TM), relative to T 1 , Eq. [2] is We assume that the net transverse magnetization immediately a valid approximation, but it begins to fail as TE and/or TM before the first pulse is equal to zero. This condition may be increases, giving elevated T 1 values for increasing TE and/ easily met by allowing at least 3T * 2 for acquisition, or applying or TM. Although Eq. [1] has been previously employed as a a homospoil gradient pulse at the end of the sequence. correction factor for localized STEAM 1 H MRS concentration Over the interval t 1 -0, the longitudinal magnetization measurements (8, 9), to our knowledge, specific derivation of recovers toward M 0 , while the xy component decays at a the equation and examination of PS dependence on TE or TM natural rate of 1/T 2 . The transverse magnetization is also has not been investigated. This Note presents an analytical deridephased by application of crusher gradient G 1 , such that vation of the equation, using an approach similar to that of immediately before the second pulse at t Å t 1 , Shoup, Becker, and Farrar (10), for a STEAM sequence to M z ( t 10 ) Å M 0 [1 0 exp(0t 1 /T 1 )] evaluate the effects of TE on the T 1 measurement of the progres-M xy (t 10 ) Å M z (0 0 )exp(0t 1 /T 2 )