Magnetic Resonance Imaging.docx

MAGNETIC RESONANCE IMAGING A. Principles of Magnetic Resonance Imaging Magnetic resonance imaging* (MRI) is a noninvasive examination technique that provides anatomic and physiologic information. Similar to computed tomography (CT), MRI is a computerbased cross-sectional imaging modality. The physical principles of MRI are totally different from those of CT and conventional radiography. MRI creates images of structures through the interactions of magnetic fields and radio waves with tissues without the use of ionizing radiation. MRI was originally called nuclear magnetic resonance (NMR) imaging, with the word nuclear indicating that the nonradioactive atomic nucleus played an important role in the technique. This term was dropped because of public apprehension about nuclear energy and nuclear weapons—neither of which is associated with MRI in any way. B. Comparison of Magnetic Resonance Imaging and Conventional Radiography MRI provides cross-sectional images and serves as a useful addition to conventional x-ray techniques. On a radiograph, all body structures exposed to the x-ray beam are superimposed into one “flat” image. In many instances, multiple projections or contrast agents are required to distinguish one anatomic structure or organ clearly from another. Cross-sectional imaging techniques such as ultrasonography, CT, and MRI more easily separate the various organs because there is no superimposition of structures. Multiple slices (cross sections) or three-dimensional volumes are typically required to cover a single area of the body. In addition to problems with overlapping structures, conventional radiography is limited in its ability to distinguish types of tissue. In radiographic techniques, contrast (the ability to discriminate between two different tissue densities) depends on differences in xray attenuation within the object and the ability of the recording medium (e.g., film or digital detectors) to detect these differences. It is difficult for radiographs to detect small attenuation changes. Typically conventional radiographs can distinguish only tissues with large differences in attenuation of the x-ray beam (air, fat, bone, and metal). Soft tissue structures such as the liver and kidneys cannot be separated by differences in x-ray attenuation alone. For these structures, differences are magnified through the use of contrast agents. However, multislice helical CT, with its superior resolving power, is much more sensitive to these small changes in x-ray attenuation and is able to distinguish the liver from the kidneys on the basis of their differing x-ray attenuation and by position. By manipulating completely different physical principles (interactions of matter with magnetic fields and radio waves), MRI is able to distinguish very small contrast differences among tissues. C. Historical Development In the mid-1940s, Felix Bloch, working at Stanford University, and Edward Purcell, working at Harvard University, discovered the principles of nuclear magnetic resonance. Their work led to the use of nuclear magnetic spectroscopy for the analysis of complex molecular structures and dynamic chemical processes. This process is still in use today for the nondestructive testing of chemical compounds. In 1952, Bloch and Purcell were jointly awarded the Nobel Prize in physics for their development of new ways and methods for nuclear magnetic precision measurements.

In 1969, Raymond Damadian proposed the first MRI body scanner. He discovered that the relaxation times of tumors differed from the relaxation times of normal tissue. This finding suggested that images of the body might be obtained by producing maps of relaxation rates. In 1973, Paul Lauterbur published the first crosssectional images of objects obtained with MRI techniques. These first images were crude, and only large objects could be distinguished. Mansfield further showed how the signals could be mathematically analyzed, which made it possible to develop useful imaging techniques. Mansfield also showed how extremely fast imaging could be achieved. Since that time, MRI technology has advanced rapidly. Very small structures are commonly imaged quickly and with increased resolution and contrast. In 2003, the Nobel Prize in physiology or medicine was jointly awarded to Lauterbur and Mansfield for their discoveries in MRI. D. Physical Principles 1. SIGNAL PRODUCTION The structure of an atom is often compared with the structure of the solar system, with the sun representing the central atomic nucleus and the planets representing the orbiting electrons. MRI uses properties of the nucleus to generate the signal that contains the information used to construct the image. Clinical MRI scanners “image” hydrogen because it is the most abundant element in the body and is the strongest nuclear magnet on a per-nucleus basis. Elements with odd atomic numbers, such as hydrogen, have magnetic properties causing them to act like tiny bar magnets (Fig. 30-1). Ordinarily, in the absence of a strong magnetic field, these protons point in random directions, as shown in Fig. 30-2, creating no net magnetization. At this point they are not useful for imaging. If the body is placed within a strong uniform magnetic field, the protons will attempt to align themselves in one of two orientations, with the field (parallel) or against the field (antiparallel). A slight majority will align with, or parallel to, the main magnetic field, also called the longitudinal plane, causing the tissues to be magnetized or have a slight net magnetization. The protons do not line up precisely with the external field but at an angle to the field causing them to rotate around the direction of the magnetic field in a manner similar to the wobbling of a spinning top. This wobbling motion, depicted in Fig. 30-3, is called precession and occurs at a specific frequency (rate) for a given atom’s nucleus in a magnetic field of a specific strength. These precessing protons can only absorb energy if that energy is presented at same frequency they are wobbling. In MRI, radiofrequency (RF) pulses at that specific precessional frequency are used. The absorption of energy by the precessing protons is referred to as resonance. This resonant frequency, called the Larmor frequency, varies depending on the field strength of the MRI scanner. At a field strength of 1.5 tesla, the frequency is approximately 63 MHz; at 1 tesla, the frequency is approximately 42 MHz; at 0.5 tesla, the frequency is approximately 21 MHz; and at 0.2 tesla, the frequency is approximately 8 MHz. When the RF pulse, at the Larmor frequency, is applied, the protons absorb the energy resulting in a reorientation of the net tissue magnetization into a plane perpendicular to the main field. This is known as the transverse plane. The protons in the transverse plane also precess at the same resonant frequency. The precessing protons (a moving magnet) in the tissues create an electrical current, the MRI signal, in the receiving coil or antenna. This follows Faraday’s law of induction, in which a moving magnetic field (hydrogen protons) induces electrical current in a coil of wire (RF antenna or RF coil).

The MRI signal is picked up by this sensitive antenna or coil, amplified, and processed by a computer to produce a sectional image of the body. This image, similar to the image produced by a CT scanner, is a digital image that is viewed on a computer monitor. Because this is a digital image, it can be manipulated, or postprocessed, to produce the most acceptable image. Additional processing can be performed on a threedimensional workstation if applicable, and hard copies can be produced if necessary. Many other odd-numbered nuclei in the body are being used in MRI. Nuclei from elements such as phosphorus and sodium may provide useful or differing diagnostic information than hydrogen nuclei, particularly in efforts to understand the metabolism of normal and abnormal tissues. Metabolic changes may prove to be more sensitive and specific in detecting abnormalities than the more physical and structural changes recognized by hydrogen-imaging MRI. Nonhydrogen nuclei may also be used for combined imaging and spectroscopy, in which small volumes of tissue may be analyzed for chemical content.

2. SIGNIFICANCE OF THE SIGNAL Conventional radiographic techniques, including CT, produce images based on a single property of tissue: x-ray attenuation or density. MR images are more complex because they contain information about differing properties of tissue—proton density, relaxation rates, and flow phenomena. Each property contributes to the overall strength of the MRI signal. Computer processing converts signal strength to shades of gray on the image. Strong signals are represented by white in the image, and weak signals are represented by black. One determinant of signal strength is the number of precessing protons in a given volume of tissue. Signal strength that depends on the concentration of protons is termed proton density. Most soft tissues, including fat, have a similar number of protons per unit volume; therefore, the use of proton density characteristics alone poorly separates these tissues. Some tissues have few hydrogen nuclei per unit of volume; examples include the cortex of bone and air in the lungs. These tissues have a weak signal as a result of low proton density and can be easily distinguished from other tissues. MRI signal intensity also depends on the relaxation times of the nuclei. Relaxation is the release of energy by the excited protons. Excited nuclei relax through two processes. The process of nuclei releasing their excess energy to the general environment or lattice

(the arrangement of atoms in a substance) is called spin-lattice relaxation. The rate of this relaxation process is measured in milliseconds and is labeled as T1. Spin-spin relaxation is the release of energy by excited nuclei through interaction among themselves. The rate of this process is also measured in milliseconds but is labeled as T2. The rates of relaxation (T1 and T2) occur at different rates in different tissues. The environment of a hydrogen nucleus in the spleen differs from that of one in the liver; therefore, their relaxation rates differ, and the MRI signals created by these nuclei differ. The different relaxation rates in the liver and spleen result in different signal intensities and appearances on the image, enabling the viewer to discriminate between the two organs. Similarly, fat can be separated from muscle, and many tissues can be distinguished from others, based on the relaxation rates of their nuclei. The most important factor in tissue discrimination is the relaxation time. The signals produced by MRI techniques contain a combination of proton density, T1, and T2 information. It is possible, however, to obtain images “weighted” toward any one of these three parameters by stimulating the nuclei with certain specific radio-wave pulse sequences. In most imaging sequences, a short T1 (fast spinlattice relaxation rate) produces a high MRI signal on T1-weighted images. Conversely, a long T2 (slow spin-spin relaxation rate) generates a high signal on T2-weighted images. The final property that influences image appearance is flow. For complex physical reasons, moving substances usually have weak MRI signals. (With some specialized pulse sequences, the reverse may be true; see the discussion of magnetic resonance angiography [MRA] later in the chapter.) With standard pulse sequences, flowing blood in vessels produces a low signal and is easily discriminated from surrounding stationary tissues without the need for the contrast agents required by regular radiographic techniques. Stagnant blood, such as an acute blood clot, typically has a high MRI signal in most imaging schemes as a result of its short T1 and long T2. The flow sequences of MRI may facilitate the assessment of vessel patency or the determination of the rate of blood flow through vessels (Fig. 30-4).

E. Equipment MRI requires a patient area (magnet room), an equipment room, and an operator’s console. A separate diagnostic workstation is optional. 1. CONSOLE The operator’s console is used to control the imaging process (Fig. 30-5). Sitting at the console allows the operator to interact with the system’s computers and electronics to manipulate all necessary examination parameters and perform the appropriate examination. Images are viewed on a computer monitor to ensure that the examination is of appropriate diagnostic quality. Images can be manipulated here, and hard copies of the exam can be produced if necessary. An independent or three-dimensional workstation may be used to perform additional imaging manipulation or post processing when required.

2. EQUIPMENT ROOM The equipment room houses all the electronics and computers necessary to complete the imaging process. The RF cabinet controls the transmission of the radiowave pulse sequences. The gradient cabinet controls the additional timevarying magnetic fields necessary to localize the MRI signal. The array processors and computers receive and process the large amount of raw data received from the patient and constructs the images the operator sees on the operator’s console. 3. MAGNET ROOM The magnet is the major component of the MRI system in the scanning room. This magnet must be large enough to surround the patient and any antennas (coils) that are required for radio-wave transmission and reception. Antennas are typically wound in the shape of a positioning device for a particular body part. These are commonly referred to as coils, or RF antennas. As the patient lies on the table, coils are either placed on, under, or around the part to be imaged. Once positioned the patient is advanced into the center of the magnet (isocenter) (Fig. 30-6). Various magnet types may be used to provide the strong uniform magnetic field required for imaging, as follows:  Resistive magnets are simple but large electromagnets consisting of coils of wire. A magnetic field is produced by passing an electrical current through the wire coils. High magnetic fields are produced by passing a large amount of current through numerous coils. The electrical resistance of the wire produces





heat and limits the maximum magnetic field strength of resistive magnets. The heat produced is conducted away from the magnet by a cooling system. Superconductive (cryogenic) magnets are also electromagnets. Their wire loops are cooled to very low temperatures with liquid helium to reduce electrical resistance. This permits higher magnetic field strengths than produced by resistive magnets. Permanent magnets are a third source for producing the magnetic field. A permanent magnet has a constant field that does not require additional electricity or cooling. The early permanent magnets were extremely heavy even compared with the massive superconductive and resistive units. Because of their weight, these magnets were difficult to place for clinical use. With improvements in technology, permanent magnets have become more competitive with the other magnet types. The magnetic field of permanent magnets does not extend as far away from the magnet (fringe field) as do the magnetic fields of other types of magnets. Fringe fields are a problem because of their effect on nearby electronic equipment.

Various MRI systems operate at different magnetic field strengths. Magnetic field strength is measured in tesla (T) or gauss (G). Most MRI examinations are performed with field strengths ranging from 0.2 to 3 tesla. Resistive systems generally do not exceed 0.6 tesla, and permanent magnet systems do not exceed 0.3 tesla. Higher field strengths require superconductive technology, with popular field strengths of 1.5 tesla and 3 tesla. Most research has concluded that field strengths used for diagnostic clinical imaging do not produce any substantial harmful effects. Regardless of magnet type, MRI units are a challenge to install in hospitals. Current units are quite heavy—up to 10 tons for resistive and superconductive magnets and approximately 100 tons for some permanent magnets. Some institutional structures cannot support these weights without reinforcement. In addition, choosing a location for the MRI unit can be difficult because of magnetic fringe fields. With resistive and superconductive magnets, the fringe field extends in all directions and may interfere with nearby electronic or computer equipment, such as television monitors and other electronic devices. In addition, metal objects moving near the magnetic fringe field, such as automobiles or elevators, may cause ripples in the field, similar to the ripples caused by a pebble thrown into a pond. These ripples can be carried into the center of the magnet, where they distort the field and ruin the images. Efforts are made to shield the magnetic fringe field to prevent its extension beyond the MRI suite. Shielding will limit the effects of the magnetic field on metal objects or electronic devices and their effect on the magnetic field.

Stray radio waves present another difficulty in the placement of MRI units. The radio waves used in MRI may be the same as the radio waves used for other nearby radio applications. Stray radio waves can be picked up by the MRI antenna coils and interfere with normal image production. MRI facilities require specially constructed rooms to shield the receiving antennas from outside radio interference, adding to the cost of the installation. Specialty units have become available for limited applications. One example is an extremity MRI scanner (Fig. 30-7). This unit is designed so that the patient can sit comfortably in a chair while having an extremity or musculoskeletal joint imaged. These units are lightweight (approximately 1500 lb) and take up less space than conventional MRI scanners, and they produce good image quality (Fig. 30-8)