Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Introduction and Physics of Body MRI
▪ Magnetic Resonance Imaging: What is the Objective?
Magnetic resonance imaging (MRI) exploits the inherent magnetism of the protons that constitute the human body in a creative way—through manipulation with radiofrequency (Rf) energy in the presence of a strong magnetic field. This manipulation induces the protons to emit energy, which is detected and reconstructed into an image. The human body—not ostensibly magnetic—is effectively magnetized by a strong magnet. Once magnetized, Rf energy shifts magnetized protons to a higher energy state. Subsequently, the protons release this energy in the process of returning to their original low-energy state. The released energy is detected in a specialized receiver (referred to as a coil in MRI parlance). With this information, ultimately images with spatial and molecular information are reconstructed without harmful effects to the patient (such as ionizing radiation).
▪ Magnetism: How is the Human Body Magnetized?
The process of magnetizing the human body actually involves only select magnetically active nuclei ( Table 1.1 ). The term, magnetically active nuclei, refers to those nuclei with unpaired protons or neutrons. These magnetically active nuclei harbor a net charge—the requisite property for interaction with a magnetic field (although neutrons have no actual net charge, the distribution of component charges is not uniform).
TABLE 1.1
Biologically Relevant Nuclei
| Element | Protons | Neutrons | Nuclear Spin | Gyromagnetic Ratio (MHz/T) |
Natural Abundance (%) | Angular Momentum (MHz) |
|---|---|---|---|---|---|---|
| 1 H (Protium) | 1 | 0 | ½ | 42.5774 | 99.985 | 63.8646 |
| 13 C | 6 | 7 | ½ | 10.7084 | 1.10 | 16.0621 |
| 15 N | 7 | 8 | ½ | 4.3173 | 0.366 | 6.4759 |
| 17 O | 8 | 9 | 5⁄2 | 5.7743 | 0.038 | 8.6614 |
| 19 F | 9 | 10 | ½ | 40.052 | 100 | 60.078 |
| 23 Na | 11 | 12 | 3⁄2 | 11.2686 | 100 | 16.9029 |
| 31 P | 15 | 16 | ½ | 17.2514 | 100 | 25.8771 |
This interaction involves two phenomena—magnetic alignment and spin, or angular momentum. Magnetic alignment describes the tendency of the magnetically active nucleus (or “magnetic moment,” or “spin”)—a miniature magnet itself—to align along the orientation of an external magnetic field ( Fig. 1.1 ). The alignment of these magnetic moments is quantized into one of two energy states: 1) parallel to (or “spin up”) or 2) anti-parallel to (or “spin down”) to the magnetic field.
The second phenomenon— spin , or angular momentum—describes the propensity of a nucleus with a net charge to oscillate like a gyroscope (or “precess”) in the presence of a magnetic field ( Fig. 1.2 ). The rate of precession is nucleus-specific and defined by a variable known as the gyromagnetic ratio (γ) .
Resonance capitalizes on nuclear precession in MRI. Energy absorption by a precessing nucleus exposed to oscillating energy of equal frequency defines resonance . By altering the oscillating frequency, only specific nuclei are selected and energized, establishing the spectroscopic basis of MRI.
MRI is founded on these two nuclear phenomena—spin and magnetic moment—occurring only in nuclei with a net charge and applicable only to few nuclei in the human body (see Table 1.1 ). Among the biologically occurring magnetically active nuclei, it is the hydrogen nucleus ( 1 H) that serves as the substrate for MRI because of its large magnetic moment (proportional to the magnetic resonance [MR] “signal,” or emitted energy converted to visual images) and abundance in the human body (ie, fat and water molecules).
▪ The Components
The Magnet
The heart of the MRI apparatus is the magnet, or main magnetic field—referred to as B 0 . Without a strong external magnetic field, the body’s protons align themselves randomly, yielding no net magnetization and, therefore no potential signal to convert to an image when subjected to Rf energy. The vector sum of the randomly aligned proton magnetic poles is zero—they cancel each other out. In the presence of a strong magnetic field—B 0 —protons align themselves parallel and anti-parallel to the magnetic field (see Fig. 1.1 ). Because more protons align parallel versus anti-parallel to the magnetic field, a net magnetic vector (NMV) is created from the protons in an external magnetic field ( Fig. 1.3 ). This magnetic vector, representing net magnetism, is the basis for creating MR images—the sine qua non of MRI. Creating this magnetism explains the need for a strong magnetic field. The stronger the magnetic field, the greater the discrepancy between parallel and anti-parallel spins, with fewer aligning anti-parallel in the higher energy state. The result is a larger NMV—the currency used to fashion MR images.
Whereas different types of commercially used magnets exist, the superconducting type is most clinically relevant to body MRI. Body MRI applications require high magnetic field strength (at least 1.0 Tesla and optimally 1.5 Tesla) in order to image rapidly with adequate signal-to-noise ratio (SNR). Resistive magnets top out at 0.5 Tesla and permanent magnets are also generally manufactured at lower magnetic field strengths (≤0.7 Tesla) not optimized for body MRI applications. Conceptually, the superconducting magnet is a large solenoid composed of superconducting wire (ie, niobium-titanium or niobium-tin), which is supercooled (with liquid helium or nitrogen) ( Fig. 1.4 ). Superconducting wire that is cooled appropriately permits the flow of electric current with virtually no resistance. By virtue of the thumb rule (officially Ampere’s Law), the result is a magnetic field oriented along the axis of the solenoid (B 0 ).
Rf System
Another key component of the MRI apparatus is the Rf transmitter system that generates the Rf pulse exciting the magnetized protons. Four components constitute the Rf transmitter system: the frequency synthesizer, the digital envelope of Rfs, a high-power amplifier, and an antenna in the form of a “coil.” The net effect is generation of an Rf pulse to excite the magnetized protons by exploiting MR.
In order to explain this process and the concept of MR, a basic understanding of nuclear physics and magnetism is necessary. As mentioned earlier, protons in a magnetic field become aligned, and the body becomes magnetized. In addition to aligning parallel or anti-parallel to B 0 , the protons rotate—or precess—around their magnetic axis, referred to as magnetic spin (see Fig. 1.2 ). The angular momentum (ω 0 ) and, accordingly, the frequency of precession (f 0 ) vary according to the strength of the magnetic field (B 0 ) and the gyromagnetic ratio (γ), which is a function of the specific properties of the nucleus—expressed by the Larmor equation:
which simplifies to
Magnetic spin precessing at the frequency of the Rf pulse will absorb energy and move to the higher energy state. Thereafter, excited protons “relax,” emitting the absorbed energy and returning to their original low-energy state. The emitted Rf energy constitutes the signal that ultimately generates an MR image. Conceptually, the NMV is aligned parallel to B 0 preceding the Rf excitation pulse. The Rf excitation pulse shifts spins into the higher energy state and the NMV away from the longitudinal axis of B 0 into the transverse plane. So, initially, the NMV is longitudinal—parallel to B 0 —and tilted by the Rf excitation pulse away from B 0 into the transverse plane ( Fig. 1.5 ). The transverse component of the spin vector ultimately constitutes the MR signal.
The Gradient System
A gradient system (a spatially varying magnetic field superimposed on spatially uniform B 0 ) distorts the magnetic environment in order to selectively excite a region—or slice—of tissue at a time to facilitate image generation and to send spatial information into the excited volume of protons. The gradient system includes three separate gradients each designed for its designated orthogonal plane: x, y, and z ( Fig. 1.6 ). Each gradient is a coil through which current passes to induce changes in B 0 and a linear variation in the main magnetic field (B 0 ) along its respective axis. In other words, a gradient alters the B 0 along a scale such that the magnetic field strength at one end of the gradient is stronger than the other (see Fig. 1.6 ).
The z—or slice-select—gradient (G z or G s ) establishes the environment in which a specific slice of protons is excited. By varying the magnetic field strength along the axis of B 0 , the slice-select gradient concordantly varies the precessional frequency of the protons along the B 0 axis. Consequently, an Rf pulse emitted with a narrow range of frequencies excites only a thin slice of protons ( Fig. 1.7 ). The narrow range of frequencies included in the Rf pulse—the transmit bandwidth—thereby determines the thickness of the excited slice of protons. This slice of excited protons ultimately constitutes the MR image.
X- and y-gradients incorporate additional spatial information into the excited slice of protons, allowing the emitted MR energy to be converted into an MR image. The gradients are applied in axes orthogonal to the slice-select gradient. The x-gradient—or frequency-encoding gradient or readout gradient (G x or G f )—applied perpendicular to B 0 —functions analogously to the slice-select gradient. By orchestrating a gradient magnetic field, spins vary in precessional frequency along a spectrum from one end of the excited slice of protons to the other ( Fig. 1.8 ). Because spins precessing at different frequencies result in destructive interference, which reduces the emitted signal, the frequency-encoding gradient is applied in two separate phases, or lobes—the dephasing and rephasing lobes ( Fig. 1.9 ).
The y-gradient—or phase-encoding gradient (G y or G p )—encodes spatial information into the excited slice of protons along the final orthogonal axis. Applied briefly, the phase-encoding gradient induces a magnetic field gradient along the final orthogonal axis such that spins at one end transiently spin faster than spins at the opposite end ( Fig. 1.10 ). Thereafter, when turned off, the spins retain their differential phase varying across the phase-encoding direction. This phase variation constitutes the spatial information along the phase-encoding axis, which is incorporated into the emitted resonance signal.
This complex sequence of Rf energy and magnetic field gradients is precisely timed to accomplish the feat of coaxing a coherent emission of resonance energy from a specific slice or volume of protons that can be received by a specialized antenna, or coil ( Fig. 1.11 ). Variations on this basic theme constitute the different pulse sequences used in MRI—such as spin-echo (SE), fast spin-echo (FSE), single-shot fast spin-echo (SSFSE), gradient-echo (GE), steady-state free precession (SSFP), and echo planar imaging (EPI). For a brief discussion of different pulse sequences, refer to the upcoming section in this chapter; for more detail on this subject, refer to texts dedicated to MRI physics— MRI Principles, D. G. Mitchell; MRI Basic Principles and Applications, M. A. Brown and R. C. Semelka; The MRI Manual, R. B. Lufkin; MRI: The Basics, R. H. Hashemi, W. G. Bradley Jr., and C. J. Lisanti; and MRI in Practice, C. Westbrook and C. Kaut.
You're Reading a Preview
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here
