Error reduction and parameter optimization of the TAPIR method for fast T₁ mapping

Abstract

A methodology is presented for the reduction of both systematic and random errors in T 1 determination using TAPIR, a Look-Locker-based fast T 1 mapping technique. The relations between various sequence parameters were carefully investigated in order to develop recipes for choosing optimal sequence parameters. Theoretical predictions for the optimal flip angle were verified experimentally. Inversion pulse imperfections were identified as the main source of systematic errors in T 1 determination with TAPIR. An effective remedy is demonstrated which includes extension of the measurement protocol to include a special sequence for mapping the inversion efficiency itself. Measurement of the spin-lattice relaxation time constant, T 1 , using MRI is becoming increasingly popular in such different applications as quantitative tracer kinetics (1-3), radiation dosimetry (4), and the response of tumors to therapy (5). The recent increasing interest in T 1 mapping in vivo is largely associated with scanner hardware improvements and the development of faster, more accurate imaging methods. Thus, a number of techniques which were previously either too demanding on the hardware or which required unacceptably long acquisition times have now developed into clinically useful methods (6 -11). Recent developments in the field have been reviewed by Kingsley in an annotated bibliography (12).In an effort to circumvent the major limitations of current methodology, we have developed a technique, based on the Look-Locker approach, and have demonstrated its potential (13-15). The method, termed TAPIR (T 1 mapping with partial inversion recovery), uses a magnetization-preparation scheme similar to that employed by Deichmann et al. (6). Additionally, TAPIR utilizes readout gradient reversals to acquire multiple lines of k-space following a single excitation pulse. Subsequently, the acquired data are reordered to fill k-space bands (segments). Interleaving slices and time-points on the recovery curve affords the acquisition of multiple slices with practically no additional time penalty. The TAPIR pulse sequence diagram is presented in Fig. 1 and is described fully in the figure caption. The multislice, multitime-point data acquisition capabilities inherent to TAPIR are in contrast to many of the previous techniques. The principle of interleaving slices and time points is extremely efficient in terms of data collection and enables the recovery curve to be sampled with a higher apparent temporal resolution than the IR-EPI technique, snapshot FLASH method, or the segmented snapshot FLASH technique. The overall speed of the sequence is enhanced not only by the rapid repetition of the sampling RF pulses, but also by minimizing the dead time and by the acquisition of multiple echoes following each excitation pulse. The relatively long recovery time for in vivo applications is used to facilitate multislice acquisition, which confers the advantage that whole-brain coverage can be achieved, as desired for human brain segmentation, for instance.The u...

Quantitative measurement of blood‐brain barrier permeability in human using dynamic contrast‐enhanced MRI with fast T1 mapping

“…The imaging protocol consisted of three anatomical sequences, a Gd-DTPA contrast-enhanced scan, and a blood flow measurement (using bolus tracking and not reported here). Before T 1 -mapping studies, (28). This parameter selection resulted in a resolution of 1.718 mm/pixel, and a sampling interval (between starts of successive TAPIR acquisitions) of 3 min 30 s. The first TAPIR T 1 data set was acquired before Gd-DTPA injection with an automatic power injector (Medrad Spectris Solaris ® MR injection system).…”

Measurement of kinetic parameters in skeletal muscle by magnetic resonance imaging with an intravascular agent

“…A multislice variant of the Look-Locker sequence, termed T 1 mapping with partial inversion recovery (TA-PIR), was used to measure pre-and postcontrast relaxation rates. During a TAPIR acquisition, the signal is acquired repeatedly during T 1 recovery, which results in an apparent relaxation time, T 1 *, that is smaller than the true T 1 (20,21). The relationship between the two is a function of the flip angle (␣) and the repetition time (TR) and is given by: …”

Enhancing the precision of quantitative water content mapping by optimizing sequence parameters

“…3). It is well known that the precision of T 1 increases with increasing (11). Interestingly, the separation between the Ernst angle, ␣ Ernst , and the angle at which the minimum of WMR is achieved (␣ Min ), also increases with increasing .…”

“…The following parameters were employed: TR ϭ 15 ms; TE 1 /TE 2 /TE 3 ϭ 2.8/5.1/7.5 ms; ␣ ϭ 25°; sequential excitation; delay time ϭ 2 s. Twenty time points were sampled on the T 1 relaxation curve and inefficiencies of the inversion pulse were determined and corrected using the procedure described in Ref. 11. The same slices with the same thickness and FOV used for the QUTE acquisition were acquired to allow for simple coregistration of the resulting images without additional image processing.…”