## A T1 and Timeto Repeat

Get Instant Access

T1 relaxation may be visualized as follows: At the conclusion of an excitation RF pulse, the magnetization is situated away from its equilibrium state. It takes a finite amount of time for the magnetization to recover to its equilibrium configuration. This recovery time is specified with a characteristic time constant that is known as T1. A series of image acquisitions produces signal intensity that is dependent on how rapidly the RF pulses are repeated relative to the rate at which the system is capable of relaxing. It is conventional in MRI to specify the rate of repetition with a parameter called time-to-repeat (TR). Figure 10 illustrates that repeated acquisitions done at short TR tend to produce a relatively weak signal (compared to longer TR) because the magnetization does not have sufficient time to reattain its equilibrium configuration. A series of measurements in which signal intensity is measured as TR is increased typically leads to results illustrated in Fig. 10. Increasing signal is observed as TR is lengthened and this generally follows an exponential mathematical function. Such a mathematical function can be completely described by a single ''time constant,'' defined as the time at which some fraction (usually 50 or 63%) of the process is complete. Therefore, the relaxation time constant, T1, is defined as the TR at which 63% of the total available signal is measured. In other words, the T1 value is determined by locating the point along the curve where 63% of the total available signal is obtained and then projecting to the TR axis.

As indicated previously, the actual value of T1 depends on the extent to which the tissue ultrastructure constrains the molecular movement. Figure 11 illustrates general features of T1 relaxation in unique brain tissues. Tissues such as CSF in which the motion of water molecules is relatively unconstrained (and therefore rapid) tend to produce relatively long T1 values. T1 values in GM or WM are smaller than in CSF because the tissue ultrastructure constrains the water movement to a greater extent. Motional constants in WM are somewhat greater than in GM due to the presence of rigid myelin structures; WM T1 values are therefore shorter compared to GM T1 values. It is indeed fortuitous that the tissue T1 values are what they are. The TR must be on the order of T1 to realize an appreciable fraction of the total available signal. Brain T1 values are such that brain MRI may be performed with TR between 500 and 2000 msec. Therefore, a typical image acquisition (128 phase encodes) requires approximately 2 min. Were the brain T1 values to be significantly greater (as is the case for distilled water) the same acquisition would take 10 times longer.

### B. T1-Weighted Imaging

Figure 11 illustrates the appearance of a T1-weighted (T1w) image. By definition, a T1w image is acquired using repetitive measures (for phase encoding and signal averaging) with full excitation (90° RF pulses) and TR that is between about 300 and 1000 msec. Image acquisition at this TR maximizes signal contrast between tissue types (Fig. 11). CSF shows the lowest (nearest to black on the gray scale) signal intensity. GM shows a somewhat stronger signal intensity (closer to white on the gray scale), but this is appreciably weaker than that produced by WM. The fatty tissue (not plotted on the graph) located around the orbits and in the superficial soft tissues shows a

Figure 10 Longitudinal (T1) recovery and its relationship to the time-to-repeat (TR).
Figure 11 Contrast development through Tl-weighted imaging.

very rapid T1 compared to brain tissues and this appears as intense signal in fatty tissues. The signal intensity differences in Tlw imaging permit the reader to readily distinguish different tissue types and, thereby, to visualize the details of anatomic structure with great clarity.

Relaxation may be further enhanced by the use of relaxation agents, which are also called paramagnetic contrast agents. These materials have inherent magnetic properties arising from unpaired electron spins. They weakly associate with water molecules and through transient contact relax the water proton nuclear spins, making T1 very short. Intravenously infused paramagnetic contrast agents are commonly employed for clinical neuroimaging. The blood-brain barrier limits the access that such materials may have to normal brain tissue, and T1 of the normal brain tissue inside the intact blood-brain barrier is not altered. On the other hand, the agents tend to readily penetrate into tissue in which the blood-brain barrier is not intact (e.g., tumors). This leads to a relatively short T1 in tissues having a damaged blood-brain barrier and to very intense signal on T1w images. The image shown in Fig. 11 was obtained after intravenous administration of a commercially available contrast agent. It illustrates an intense double ring enhancement pattern produced by a tumor located at the right

(viewer's left) frontotemporal junction. The bright "ring" is generally thought to represent highly vascular living tumor tissue that does not have an intact blood-brain barrier.

Figure 12 illustrates that T1w imaging may also be used to obtain angiographic images showing the major intracerebral vessels. This is known as magnetic resonance angiography (MRA). Blood flow tends to accentuate T1 relaxation because flow is constantly moving new water into the tissue section that is being imaged. Therefore, intravascular water tends to show full signal intensity after each excitation because it was not in the slice during previous excitations. On the other hand, the brain water needs time to relax between excitations to produce strong signal. One MRA approach is to employ very heavily T1w contrast (i.e., very short TR) so that brain signal is very weak and the intravasular water produces a much stronger signal. This is illustrated in the image shown on the left in Fig. 12. One can appreciate spots and strings of high signal intensity that depict the major vessels in this heavily T1w (flow-sensitized) image. Although single-slice views of vascular anatomy such as this are useful, display of MRA results in a manner consistent with fluoroscopic angiography is typically employed. This is illustrated in the image shown on the right in Fig. 12. To construct an angiographic projection view,