Image weighting and contrast

2
Image weighting and contrast




After reading this chapter, you will be able to:



  • Differentiate between intrinsic and extrinsic contrast parameters.
  • Explain T1 recovery and T2 decay.
  • Understand how contrast is generated in different tissues.
  • Apply what you have learned to create images of different weighting.
  • Describe techniques that affect image contrast.

INTRODUCTION


All clinical diagnostic images must demonstrate contrast between normal anatomical features and between anatomy and pathology. If there is no contrast, it is impossible to identify anatomy or detect abnormalities within the body. One of the main advantages of MRI compared to other imaging modalities is its excellent soft tissue discrimination. The contrast characteristics of each image depend on many variables, and it is important that the mechanisms that affect image contrast in MRI are clearly understood.


IMAGE CONTRAST


The factors that affect image contrast in diagnostic imaging are usually divided into two categories.



  • Intrinsic contrast parameters are those that cannot be changed because they are inherent to the body’s tissues.
  • Extrinsic contrast parameters are those that can be changed because they are under our control.

For example, in radiography, intrinsic contrast parameters include the density of structures through which the X-ray beam passes and is attenuated by, while extrinsic contrast parameters include the exposure factors. These determine image contrast in a radiograph, but exposure factors can be changed, whereas tissue density cannot. In MRI, there are several parameters in each group.


Intrinsic contrast parameters are as follows:



  • T1 recovery time
  • T2 decay time
  • Proton density (PD)
  • Flow
  • Apparent diffusion coefficient (ADC).

All these are inherent to the body’s tissues and cannot be changed. T1 recovery time, T2 decay time, proton density, and ADC are discussed in this chapter. Flow is discussed in Chapter 8.


Extrinsic contrast parameters are as follows:



  • TR
  • TE
  • Flip angle
  • TI
  • Turbo factor/echo train length
  • b value.

These are all selected in the scan protocol. Some of these parameters depend on the pulse sequence we choose (see Table 4.18). TR, TE, and the b value are discussed later in this chapter, and the other parameters are described in Chapters 3 and 4 in the relevant pulse sequence sections. A list of acronyms of the five main system manufacturers is provided at the beginning of the book. This includes the contrast parameters and pulse sequences described in these chapters.


RELAXATION


At the end of Chapter 1, we explored the consequences of switching off the radio frequency (RF) excitation pulse. To recap, as soon as the B1 field is removed, hydrogen nuclei are only under the influence of B0. One of the principles of thermodynamics is that a system always seeks its lowest possible energy level. This occurs in MRI when the RF excitation pulse is switched off, and therefore hydrogen nuclei return to their low-energy state and their magnetic moments dephase [1]. The process by which this occurs is called relaxation.


During relaxation, hydrogen nuclei give up absorbed RF energy, and the net magnetic vector (NMV) returns to B0. At the same time but independently, magnetic moments of hydrogen nuclei lose phase coherence. Relaxation therefore results in recovery of magnetization in the longitudinal plane and decay of coherent magnetization in the transverse plane.



  • The recovery of longitudinal magnetization is caused by a process termed T1 recovery.
  • The decay of coherent transverse magnetization is caused by a process termed T2 decay.

T1 RECOVERY


T1 recovery is caused by hydrogen nuclei giving up their energy to the surrounding environment or molecular lattice. The term recovery refers to the recovery of longitudinal magnetization, and T1 relates to the fact that it is the primary relaxation process. (It is not, however, the first process that occurs. The learning tip above shows that T1 recovery takes 5–10 times longer than T2 decay.) This type of relaxation is called spin–lattice energy transfer. Energy released by spins to the surrounding molecular lattice causes magnetic moments of hydrogen nuclei to recover their longitudinal magnetization. According to quantum theory, the number of high-energy spins decreases, and the number of low-energy spins increases as energy is lost by high-energy spins during the relaxation process. According to classical theory, the NMV gradually realigns itself in the longitudinal plane as the proportion of spin-up and spin-down hydrogen nuclei changes.


The rate of T1 recovery is an exponential process and occurs at different rates in different tissues (Table 2.1). As illustrated in Figure 2.1, longitudinal magnetization is related exponentially to recovery time. This means that most longitudinal recovery happens at the beginning of the time frame. As time progresses, gradually less and less longitudinal recovery occurs until the longitudinal magnetization is fully recovered. There is a time constant associated with this exponential relationship. This is called the T1 recovery time and is the time it takes for 63% of the longitudinal magnetization to recover in a tissue (Equation (2.1)) (Figure 2.1). The T1 recovery time of a tissue is an intrinsic contrast parameter that is inherent to the tissue. The time during which T1 recovery occurs is the time between one RF excitation pulse and the next. This is the repetition time (TR) (see Chapter 1). The TR therefore determines how much T1 recovery occurs in a tissue. It is therefore the variable shown on the horizontal axis of Figure 2.1.


Table 2.1 Typical T1 recovery times of brain tissue at 1 T.



















Tissue T1 recovery time (ms)
Water 2500
Fat 200
CSF 2000
White matter 500
Graph shows time versus longitudinal magnetization where one line starts from point of origin, increases and intersects imaginary horizontal and vertical line.

Figure 2.1 The T1 recovery curve.











Equation 2.1

Mzt = Mz (1 − et/T1)


Therefore


SI = (1 − et/T1)


Mzt is the amount of longitudinal magnetization at time t (ms) after the removal of the excitation pulse


Mz is full longitudinal magnetization


T1 is the T1 recovery time (ms) and is the time taken to increase the longitudinal magnetization by a
factor of e.


SI is the signal intensity in a tissue


This equation plots the size of the recovering NMV as a function of time after the removal of the excitation pulse and the T1 recovery time. When t = T1, 63% of the longitudinal magnetization recovers. When t = 2 × T1, 86% recovers and when t = 3 × T1, 95% recovers. It usually takes between 3 and 5 T1 recovery times for full recovery to occur


T2 DECAY


T2 decay is caused by the magnetic fields of neighboring hydrogen nuclei interacting with each other. The term decay refers to the loss of coherent transverse magnetization, and T2 relates to the fact that it is the secondary relaxation process. This type of relaxation is termed spin–spin relaxation and causes dephasing of magnetic moments of the spins. Spin–spin relaxation is caused by one spin transferring energy to another spin rather than into the lattice. It occurs because hydrogen nuclei are in the same environment and experiencing the same B0 field [2]. Magnetic moments of all the hydrogen nuclei (spin-up and spin-down) lose phase coherence in this way.


Imagine two spins close to each other, one with its magnetic moment aligned in the same direction as B0 and the other in the opposite direction. The spin whose magnetic moment is aligned in the same direction as B0 creates a slightly larger magnetic field than is experienced by the neighboring spin. As a result, the precessional frequency of the magnetic moment of this spin increases. Conversely, the spin whose magnetic moment is aligned in the opposite direction to B0 causes a slightly lower magnetic field than is experienced by the other spin, and its precessional frequency decreases [3]. These small changes in frequency are sufficient to cause dephasing of magnetic moments of the spins.


Spin–spin interaction is inherent to the tissue, but dephasing is also caused by inhomogeneities in the B0 field. Inhomogeneities are areas within the magnetic field that do not exactly match the external magnetic field strength. Some areas have a magnetic field strength slightly less than the main magnetic field (shown in purple in Figure 2.2), while other areas have a magnetic field strength slightly higher than the main magnetic field (shown in red in Figure 2.2).

Diagram shows two circles labeled homogenous field on left and inhomogeneous field with keys for lower than center frequency, center frequency, and higher than center frequency.

Figure 2.2 T2* decay and field inhomogeneities.


According to the Larmor equation, the Larmor frequency of an MR-active nucleus is proportional to the magnetic field strength it experiences. If a hydrogen nucleus lies in an area of inhomogeneity with higher field strength, the precessional frequency of its magnetic moment increases, i.e. it speeds up. However, if a hydrogen nucleus lies in an area of inhomogeneity with lower field strength, the precessional frequency of its magnetic moment decreases, i.e. it slows down. This relative acceleration and deceleration of magnetic moments due to magnetic field inhomogeneities, and differences in the precessional frequency in certain tissues, causes immediate dephasing of the magnetic moments of the spins and produces a free induction decay (FID) as shown in Figures 2.2 and 2.3.

Diagram shows four circles labeled T2 dephrasing on top and graph shows corresponding signal loss due to free induction decay where line increases and decreases and reduces slowly.

Figure 2.3 Dephasing and free induction decay.


The rate of T2 decay is an exponential process and occurs at different rates in different tissues (Table 2.2). As Figure 2.4 illustrates, coherent transverse magnetization is related exponentially to decay time. This means that there is more coherent transverse magnetization at the beginning of the time-frame and, as time progresses, there is less coherent transverse magnetization until all the magnetic moments dephase. There is a time constant associated with this exponential relationship. It is called the T2 decay time and is the time it takes for 63% of the transverse magnetization to dephase (37% is left in phase) in a tissue (Equation (2.2)) (Figure 2.4).










Equation 2.2

Mxyt = Mxy et/T2


therefore


SI = et/T2


Mxyt is the amount of transverse magnetization at time t (ms) after the removal of the excitation pulse


Mxy is full transverse magnetization


T2 is the T2 decay time (in ms) and is the time taken to reduce the transverse magnetization by a factor of e


SI is the signal intensity in a tissue


This equation plots the size of the decaying transverse magnetization as a function of time after the removal of the excitation pulse and the T2 decay time. When t = T2, 63% of the coherent transverse magnetization has decayed and 37% remains


Table 2.2 Typical T2 decay times of brain tissue at 1 T.



















Tissue T2 decay time (ms)
Water 2500
Fat 100
CSF 300
White matter 100
Graph shows time versus coherent transverse magnetization where one line starts from 100 percent on coherent transverse magnetization, decreases, and intersects imaginary horizontal and vertical line.

Figure 2.4 The T2 decay curve.


The T2 decay time of a tissue is an intrinsic contrast parameter that is inherent to the tissue. The time during which this occurs is the time between an RF excitation pulse and when signal is collected in the receiver coil (see Chapter 1). The echo time (TE) therefore determines how much T2 decay occurs in a tissue when signal is collected. It is therefore the variable shown on the horizontal axis of Figure 2.4.


Dephasing caused by inhomogeneities in the B0 field produces its own decay curve. This is differentiated from T2 decay by using the term T2*. When the RF excitation pulse is switched off, magnetic moments of the hydrogen nuclei dephase very quickly (within about 10 ms), and this is caused by T2* decay. T2 decay, from inherent tissue dephasing, takes longer than T2* decay (Equation (2.3)). The purpose of pulse sequences is to refocus or rephase magnetic moments of the hydrogen nuclei so that inherent tissue dephasing is measured at time TE and images of different contrast can be acquired (see Chapter 3).










Equation 2.3

1/T2* = 1/T2 + 1/2γΔβ0


T2 and T2* are the tissues T2 and T2* relaxation times (ms)


γ is the gyromagnetic ratio (MHz/T)


ΔB0 is the variation in magnetic field (parts per million, ppm)

This equation shows how T2 and T2* are related to each other. Poor field inhomogeneities result in T2* being much shorter than T2, and a fast decaying signal

Table 2.3 Things to remember – relaxation.











Relaxation is a general term that refers to a loss of energy. In MRI, this is energy that is delivered to the spins via the RF excitation pulse and then lost once it is switched off
Spin lattice energy transfer is a relaxation process where spins give up the energy absorbed through RF excitation to the surrounding molecular lattice of the tissue. It causes T1 recovery
T2 decay is an irreversible loss of phase coherence due to spin–spin interactions on an atomic and molecular level. It is one of the causes of T2 decay
Pulse sequences are mechanisms that permit refocusing of spins so that images can be acquired with different types of contrast

CONTRAST MECHANISMS


An MR image has contrast if there are areas of high signal (hyperintensity – white in the image) and areas of low signal (hypointensity – black in the image). Some areas have an intermediate signal (shades of gray in between white and black). The NMV is separated into individual vectors of the tissues such as fat, cerebrospinal fluid (CSF), and muscle.


A tissue has a high signal if it has a large transverse component of coherent magnetization at time TE. If there is a large component of coherent transverse magnetization, the amplitude of signal received by the coil is large, resulting in a hyperintense area on the image. A tissue returns a low signal if it has a small or no transverse component of coherent magnetization at time TE. If there is a small or no component of transverse coherent magnetization, the amplitude of signal received by the coil is small, resulting in a hypointense area on the image.


Images obtain contrast mainly through the mechanisms of T1 recovery, T2 decay, and proton or spin density. The proton density (PD) of a tissue is the number of mobile hydrogen protons per unit volume of that tissue. The higher the proton density of a tissue, the more signal available from that tissue. T1 and T2 relaxation depend on two factors:



  • If the molecular tumbling rate matches the Larmor frequency of hydrogen. If there is a good match between the rate of molecular tumbling and the Larmor frequency of magnetic moments of hydrogen, energy exchange between hydrogen nuclei and the molecular lattice is efficient. When there is a bad match, energy exchange is not as efficient. This is important in both T1 recovery and T2 decay processes.
  • If the molecules are closely packed together. In tissues where molecules are closely spaced, there is more efficient interaction between the magnetic fields of neighboring hydrogen nuclei. The reverse is true when molecules are spaced apart. This is especially important in T2 decay processes, which rely on the efficiency of interactions between the magnetic fields of neighboring hydrogen nuclei (spin–spin relaxation).

RELAXATION IN DIFFERENT TISSUES


As discussed earlier, T1 recovery and T2 decay are exponential processes with time constants T1 recovery time and T2 decay time, which represent the time it takes for 63% of the total magnetization to recover in the longitudinal plane due to spin–lattice energy transfer (T1 recovery time), or lost in the transverse plane via spin–spin relaxation (T2 decay time). Generally, the two extremes of contrast in MRI are fat and water (Figure 2.5). (In this book, fat vectors are drawn in yellow and water vectors in blue.) Let’s explore how contrast is generated in these tissues.

Graph shows transverse components of magnetization versus longitudinal components of magnetization where arrow points upward and two arrows point diagonally labeled Bo, fat vector, and water vector.

Figure 2.5 The magnitude of transverse magnetization vs amplitude of signal.


Fat and water


Fat molecules contain atoms of hydrogen arranged with carbon and oxygen. They consist of large molecules called lipids that are closely packed together and whose molecular motion or tumbling rate is relatively slow. Water molecules contain two hydrogen atoms arranged with one oxygen atom (H2O). Its molecules are spaced apart, and their molecular tumbling rate is relatively fast. The oxygen in water tends to steal the electrons away from around the hydrogen nucleus. This renders it more available to the effects of the main magnetic field. In fat, the carbon does not take the electrons from around the hydrogen nucleus. They remain in an electron cloud, protecting the nucleus from the effects of the main field. Therefore, hydrogen in fat recovers more rapidly along the longitudinal axis than water and loses transverse magnetization faster than water. Subsequently, fat and water appear differently in MR images.


T1 recovery in fat


T1 recovery occurs due to hydrogen nuclei giving up their energy to the surrounding molecular lattice. Fat has a low inherent energy and easily absorbs energy into its lattice from hydrogen nuclei. The slow molecular tumbling in fat allows the T1 recovery process to be relatively rapid because the molecular tumbling rate matches the Larmor frequency. Consequently, there is efficient energy exchange from hydrogen nuclei to the surrounding molecular lattice. This means that magnetic moments of fat hydrogen nuclei quickly relax and regain their longitudinal magnetization. The NMV of fat realigns rapidly with B0, so the T1 recovery time of fat is short (Figure 2.6).

Graph shows time versus range where line starts from point of origin and increases with six plots and label for fat regains longitudinal magnetization rapidly.

Figure 2.6 T1 recovery in fat.


T1 recovery in water


T1 recovery occurs due to hydrogen nuclei giving up energy acquired from the RF excitation pulse to the surrounding lattice. Water has a high inherent energy and does not easily absorb energy into its lattice from hydrogen nuclei. In water, molecular mobility is high, resulting in less efficient T1 recovery because the molecular tumbling rate does not match the Larmor frequency and does not allow efficient energy exchange from hydrogen nuclei to the surrounding molecular lattice. Magnetic moments of water hydrogen nuclei take longer to relax and regain their longitudinal magnetization. The NMV of water takes longer to realign with B0, and so the T1 recovery time of water is long (Figure 2.7).

Graph shows time versus range where line starts from point of origin and increases with six plots and label for water regains longitudinal magnetization slowly.

Figure 2.7 T1 recovery in water.

Mar 9, 2019 | Posted by in MAGNETIC RESONANCE IMAGING | Comments Off on Image weighting and contrast
Premium Wordpress Themes by UFO Themes