Basic Physics of MR Signal and Image Generation - Advance Image Processing in Magnetic Resonance Imaging - page 16

This process of recovery of longitudinal magnetization is very important when
calculating the contrast between tissues with different T1 values and determining
the imaging method that obtains the greatest signal-to-noise ratio (SNR).
The relaxation times T1 and T 2 are determined by the molecular environment
and thus are dependent on the sample. If the sample contains chemically shifted
resonances, the nuclei in different chemical environments exhibit T1 and T 2
values characteristic of the particular environment. Knowledge of T1 and T2
values is obtained experimentally by perturbing the equilibrium magnetization in
appropriated multiple RF pulses. For pure liquids T1 = T2 and for biological

samples T2 T1. Molecules in a mobile liquid environment have T1 and T 2

values in the range of tens of hundreds of milliseconds. For tissues in the body,
the relaxation times are in the ranges 250 msec

<

< T1 < 2500 msec and 25 msec

< T2 < 250 msec and, usually, 5T 2 < = T1 < = 10T 2.

1.6.1.3 Pseudo-Relaxation: T2*

In practice, variations in the value of the magnetic field throughout the sample
also cause dephasing of the magnetic moments, mimicking the decay of transverse
magnetization caused by relaxation. It is convenient to define another relaxation
time, T2*, describing the “observed” rate of decay of the FID:

1
2

1
2

1
2

* =+ ′

(1.20)

T T T

where T2 ′ describes the decay in signal due to the magnetic field inhomogeneity:

1
2

′ = γ∆

B (1.21)

0

T

where B 0 is the extent of variation of the applied magnetic field strength over

the region occupied by the sample. This causes the nuclei in different regions of
the sample to experience slightly different magnetic fields and hence to precess
at different Larmor frequencies. The FID obtained in such a field is seen to decay
faster than that determined by T 2 (T 2*

∆

< T 2).

1.6.2 P ROTON D ENSITY

Most of the hydrogen atoms in the tissues are within water molecules; it is these
protons that we detect in an MR experiment. The term proton density (PD) simply
refers to the number of protons per unit volume and is effectively proportional
to the density of water in the tissue. Thus, for example, bone has very low proton
density, and liver has high proton density, while blood has a very high proton
density.
It is easy to see that we can identify the proton density with M 0 , the equilib-
rium magnetization.