Nuclear Magnetic Resonance Scanners As Medical Instruments

A. System Considerations

The overall operation of a NMR scanner is controlled by a computer (Fig. 3). It provides the pulse-timing information for the gradient and radio-frequency transmitter coils. It also switches on the preamplifier and the receiver circuitry during the time when data is being acquired from the nuclear spins. The data are acquired as free induction decay or, more commonly, spin echo signals. Extensive mathematical calculations, usually a two-dimensional Fourier transformation, are required to convert the FID or spin echo data into an image and the calculations are done by the computer. For permanent magnet systems no magnet power supply is required; for superconducting magnet systems the magnet power supply is needed only initially when the magnet is being energized.

The NMR signal is sufficiently weak that interfering electrical noise is a major consideration. To minimize outside electrical interference, some form of electrical screening is normally placed around the scanner or the scan room, and electrical filters are used on those circuits that could transmit outside noise to the receiver. Figure 4 shows a patient within a screened room being positioned for scanning. In Fig. 5 the patient is seen within the magnet bore in the location that permits scanning to be done.

The strength of NMR signal generated by the precess-ing spins is proportional to the degree of alignment of the spins [M0 in Eq. (2)] and to their rate of precession (YB0). Since both those quantities are proportional to field

FIGURE 3 Block diagram of an NMR scanner. (Courtesy of GE Medical Systems.)
FIGURE 4 Preparation for scanning. The superconducting magnet shown here operates at 1.5 T. (Courtesy of GE Medical Systems.)

strength, the signal increases as B^. The noise in the system can be brought to a very low level by proper design of the electronics. However, there is some electrical noise that is generated within the patient and that cannot be removed by improved circuit design. The ultimate source of this noise is the random motion of ions and charged macro-molecules within the patient's body and is of thermal origin. The noise voltage increases approximately linearly with increasing frequency.

A key parameter in determining overall image quality is the signal-to-noise ratio (SNR). The arguments just given indicate that the SNR should increase linearly as the field strength is increased. This is the basis for the use of strong magnetic fields in NMR scanners. If it is desired to achieve high-resolution proton images, thin slices and small picture elements (as will be discussed later) must be used; however, this leads to a decrease in the amount of signal available to determine the pixel brightness numbers. If this process is pushed too far, the images will become grainy because of the effects of the noise. By increasing the field strength of the magnet, the available signal is increased; this can be used to support higher resolution imaging.

FIGURE 5 NMR scan. The patient has been positioned in the center of the magnet and can be seen through a window above the operator's console. (Courtesy of GE Medical Systems.)

In MR spectroscopy, the molecules being studied are present in a concentration very low compared to that of water. As a consequence, spectroscopy signals tend to be very weak. This places an even stronger premium on field strength than does imaging. It is widely accepted that spectroscopic studies of patients are not warranted in field strengths less than about 1.5 T.

B. Magnets

The magnet is probably the most significant portion of a whole-body scanner. The magnets used vary substantially in terms of field strength, shape, and basic design. However, they all must meet certain basic requirements. One obvious requirement is that the magnet be large enough to admit a human body and produce a field strength that is intense enough to produce a strong proton NMR signal. Equally important, though not so obvios, is the requirement for high homogeneity, or uniformity of the magnetic field strength from one part of the imaging volume to another. It is essential that, unless gradient coils are being used to deliberately modulate it, the Larmor precession proceed at the same rate throughout the entire portion of the patient that is to be imaged. The homogeneity re-quirment puts strong limitations on the acceptable magnet designs. Superconducting and resistive systems are designed to provide a cylindrical symmetry. A cylindrical opening, called the room temperature bore, is available to permit placing the patient into the region of strong and highly homogeneous field.

At the present time clinical proton imaging is usually done using magnetic field strengths in the order of 0.2 to 1.5 T, although, in about 2000, commercial MRI systems operating at 3 T began to become available. Later, in the section on current trends, research scanners operating at even higher field strengths will be described. The types of magnet that have been used to produce clinical images have included permanent magnet systems, iron-core electromagnets, and resistive and superconducting mul-ticoil magnets. Each of these designs has certain advantages. However, for producing fields greater than about 0.5 T, only the superconducting systems are practical; field strengths in this range are beyond the capabilities of present-day permanent magnet materials. In resistive systems the coils are made of a conventional conductor, such as copper or aluminum. If enough current is run through them to produce whole-body sized fields much above 0.2 T, so much heat is generated in the windings that even with water cooling there is a likelihood of burning the insulation or actually melting the wire.

The phenomenon of superconductivity provides the best approach to achieving strong magnetic fields. Since 1911 it has been known that some materials when cooled to temperatures near absolute zero can conduct electricity with absolutely no electrical resistance and, therefore, no associated heating or energy loss. Electric currents, once started in a loop of superconducting wire, havebeen shown to persist for years without decreasing measurably, even though no source of electrical energy was used to maintain them. Unfortunately, the original superconducting materials that were discovered, such as lead and tin, could not be used to generate strong magnetic fields because as the current was increased beyond a rather small value the superconducting state was destroyed and the electrical resistance was restored. In the 1960s a new class of superconducting materials capable of carrying much higher current densities was discovered. During the next decade these materials, particularly in the form of niobium-titanium alloys, became the basis of a new class of high-field magnets. These were used to make NMR spectrometers for chemical research applications that were capable of generating much stronger fields than had previously been available. They were only big enough, however, to contain small, test-tube-sized samples. Small-bore, high-field systems are now available for chemical research that provide steady fields in the range of 20-25 T and with NMR proton frequencies approaching 1000 MHz (1 GHz).

By the early 1980s the Oxford Instruments Company of Oxford, England, had produced whole-body superconducting magnets capable of reaching 1.5 T. At the present time several manufactures build whole-body magnets of this type.

In 1986 a still newer class of superconducting materials was discovered, capable of maintaining their superconducting properties to temperatures much higher than the previously known materials. These may eventually have an application in MRI, perhaps by eliminating the need to immerse the coils in liquid helium. The current-handling capabilities of these new materials, however, are at present too weak to permit their use in whole-body magnets. This situation may improve after further research.

The exceptionally high homogeneity of MRI magnets is achieved in two steps: (1) during the basic coil design and (2) by the use of shim coils during operation. The basic approach is to use a set of coils about 1.5 m in diameter positioned along the z axis of the magnet (Fig. 6). The contribution of each coil to the B0 field is determined by its location along the z axis, its radius, and the number of turns of superconducting wire wound on it multiplied by the current in the coil. The z component of the resulting magnetic field can be represented as an expansion about the center of the magnet by using specific mathematical functions, the spherical harmonics. The ze-roth order of this expansion is the perfectly uniform field, Bz = constant, that is the desired field. All other terms in the expansion represent contaminating inhomogeneities

FIGURE 6 Cutaway drawing of magnet geometry. A four-coil geometry is used to produce the static field. The inner coil shown is for the transverse gradient field. (Courtesy of GE Medical Systems.)

that are undesired. The first-order expansion terms represent gradients in all three directions, dBz/dx, dBz/dy, and dBz/dz. These gradients and a large number of additional higher order error terms can be eliminated by correctly placing the proper number of ampere turns at specific locations along the axis. For example, a six-coil design commonly utilized in clinical magnets can theoretically eliminate all the contaminating spherical harmonic terms up to the 12th order. The use of these carefully calculated coil designs greatly increases the volume within the magnet over which the homogeneity specifications can be met.

There are, however, many sources of slight manufacturing errors that prevent the ideal field from being obtained. For example, the individual coils may be slightly out of round, or slightly out of position along the z axis or not oriented absolutely perpendicular to the z axis. To correct for these inevitable manufacturing tolerances, each magnet is equipped with a set of shim coils; up to a dozen or more independent coils are usually available. Each of these coils, wound on a cylindrical coil form near the inner surface of the main field coils, has a different geometry. The geometry of a given shim coil is chosen to produce a field near the magnet center that has a pattern closely approximating a single spherical harmonic. By adjusting the current in each shim coil independently, it is possible to cancel out the residual errors associated with each of the lower order harmonics. The shim coils carry much less current than the main coil windings and, therefore, may be either superconducting or resistive. Sometimes both resistive and superconducting shim coils are provided. The current settings necessary for the shim coils to produce the maximum homogeneity for a given magnet is determined at the time of magnet manufacture in a process called shimming. This process is repeated episodically, as needed, over the life of the scanner.

Homogeneity is normally specified as the maximum deviation, in parts per million, of the field within a specified diameter spherical volume (dsv) centered on the center of the room-temperature bore. The homogeneity is better, of course, for small volumes. In a typical situation the homogeneity of a shimmed magnet might be 0.1 ppm over a 10-cm dsv, 10 ppm over a 30-cm dsv, and 40 ppm over a 50-cm dsv.

The superconducting property means that such magnets can be operated in the persistent mode. Once the power supply has increased the current to the point where the desired field has been reached (this normally is done over several hours), a superconducting switch is activated, and the power supply is disconnected. As long as the windings are kept below the superconducting transition temperature, no further input of energy is required to maintain the field. Modern magnets, operating in the persistent mode, have no trouble meeting a drift specification of less than 0.1 ppm/hr. This drift rate is so slow that the magnets can go for months or years without requiring additional energy input. To maintain the coils in the superconducting state, they are located within a double cryostat. The inner chamber contains the coils immersed in liquid helium at 4.2 K. The outer chamber contains liquid nitrogen at 77 K as an intermediate temperature reservoir. Every few weeks it is necessary to replace these cryogenic liquids as they boil off.

C. Gradient Coils

In the space between the main magnet coils and the patient it is necessary, for imaging purposes, to place a set of three coils, each of which is designed to produce a specific gradient in Bz, the z component of the static field B0. These coils are respectively the x, y, and z gradient coils. The currents in these coils are under the control of the computer, and they can be pulsed on and off in the proper sequence to aid in manipulating the spin system as required by the imaging technique. The fields produced by the gradient coils are much smaller than that of the main magnet. The gradient field strengths commonly used in present-day scanners range from 1 to 5 G/cm, which corresponds to 0.0001 to 0.0005 T/cm.

During the imaging session the patient experiences an intermittent series of tapping or banging noises. These sounds can vary in intensity, from rather soft to practically unbearable, depending on the magnitude of the currents involved and on the degree of acoustic damping used. The sounds originate in the gradient coils and result from the magnetic forces between the pulsed gradient currents and the strong, static field.

D. Radio-Frequency Coils

Between the gradient coils and the patient is located at least one additional coil, which serves as transmitter and/or receiver of rf energy. Imaging techniques require a series of 90° and 180° pulses at the Larmor frequency of the protons. These are the Bi fields used to excite the spin system. The coil that delivers these pulses must be designed to handle the high instantaneous voltages and currents that are required. It should also produce a magnetic field that is as uniform as possible across the desired field of view and at right angles to the static field B0. To the extent that this B1 field is not uniform, the pulses produced will produce errors in the desired angles of spin flipping.

The task of building a radio-frequency coil large enough to accommodate the human body, producing a uniform B1 field, and still capable of resonating at frequencies approaching 100 MHz has provided some technical difficulties. This is because the large size of the coil produces an inherently large inductance, which interacts with the stray self-capacitance of the coil to produce a self-resonance phenomenon that degrades the coil performance. This problem has been overcome by using designs with capacitors distributed along the length of the coil. A particularly effective design for whole-body use at high frequencies has been the "birdcage" concept developed by C. Hayes and his coworkers.

Scanners are usually provided with at least two sizes of cylindrical rf coils. One with a diameter of about 56 cm is large enough to accommodate the entire body. The other, smaller coil is about 28 cm in diameter and is designed for head imaging. Generally speaking, the closer a coil is to the region being imaged, the better SNR it will provide. For high-resolution imaging it has now become common to use coils, called surface coils, that are designed to fit more closely over the region of the body that is to be imaged. Therefore, specialized coils have been developed to image the spine, the neck, the shoulder, and so on.

E. Safety Considerations

In 1976 the U.S. Congress amended the Food, Drug, and Cosmetic Act of 1938 to apply certain restrictions on the introduction of new medical devices. In January 1980 the Food and Drug Administration (FDA) responded to the congressional action by issuing regulations that applied to the manufacturers of new medical devices and to researchers working with such devices. The regulations, analogous to those applied to the introduction of new drugs, made it necessary to develop data regarding the safety and efficacy of new devices prior to seeking approval for marketing them. The NMR scanners were the first major imaging device to be subjected to these regulations. During the 1980s several manufacturers successfully sought FDA approval for their scanners.

As additional experience was gained during the 1990s, FDA approval was granted for a variety of scanner enhancements, such as the use of higher field strengths and a wider variety of RF and gradient coils.

NMR scanners place the patient in an environment that is quite unlike that of any other medical instrument. Initially, there were several areas of concern that, with experience, have become better understood and appear not to represent a danger to patients. The area of greatest continuing concern is the interaction between the strong, static, magnetic field and ferromagnetic substances inadvertently brought into the region of the scanner. One version of this problem arises from the fringing field surrounding the magnet, which can be treacherously strong. Many objects common in hospitals (e.g., oxygen bottles, mops, fans, and hairpins) contain enough magnetic material that they can be drawn into the magnet with great force and rapidity. Such flying objects are extremely dangerous to anyone in their path. For this reason, most manufacturers and users of the scanners go to great lengths to limit access to the vicinity of the scanner. Permanent magnet systems, and some superconducting systems that have magnetic shielding around them, have smaller fringing fields and are less susceptible to this effect. A second version of this problem comes about because some patients have ferromagnetic substances implanted within their bodies. This is usually the result of a prior surgical procedure, such as the clipping of a diseased blood vessel, but in some cases iron fragments (e.g., shrapnel) have become embedded in a patient's tissues during some sort of accident or explosion. Patients are not always aware of the presence of these objects. Therefore care has to be taken before scanning to exclude those patients with possibly dangerous ferromagnetic implants. Implanted cardiac pacemakers can malfunction, or conceivably, be permanently damaged because of exposure to strong magnetic fields. Therefore patients with these devices in place are not normally candidates for NMR scanning.

Other areas of initial concern were the effects of the static field on normal tissue function; the possibility that electric fields generated by the rapidly changing gradient fields (i.e., the "dB/dt' effect) could cause nerve stimulation or irregularities in the cardiac rhythm; tissue heating associated with the rf excitation field; and possible effects on blood pressure resulting from forces of interaction between blood and the static magnetic field. In the scanners presently used, all of these effects appear to be readily tolerable, and in most cases, negligible. The FDA continues to receive and evaluate designs for more advanced scanners, and the regulatory aspects of scanner safety continue to evolve. Of particular note are guidelines approving the use of static fields up to 4.0 T and, under appropriately controlled conditions, an average heat input to the patient's body (specific absorption rate or SAR) of 4.0 W/kg. The rf heating that occurs under standard imaging conditions is comparable to that resulting from normal metabolic activity and is unnoticeable to the patient. Likewise, the possible effects of dB/dt on nerve excitation and the magneto-hydrodynamic interaction between flowing blood and the static magnetic field appear to require conditions far beyond those used in modern scanners. At the present time these concerns represent hypothetical, rather than actual, hazards of the scanning process.

III. IMAGING TECHNIQUES A. Selective Excitation

One of the key capabilities of MRI scanners is the ability to excite a single, thin slice of spins within the patient. This permits the construction of images that have the character of two-dimensional cross-sectional cuts through the patient's anatomy. NMR imaging resembles CT scanning in producing this type of anatomical image. However, MRI has a substantial advantage over CT in that it permits the location of the slice to be chosen electronically by the operator without moving the patient or any components of the scanner. In MRI, imaging planes of any orientation may be chosen, and these planes may also be moved electronically from side to side, top to bottom, or front to back through the patient's anatomy.

The key to selective excitation is carrying out the rf excitation in the presence of a gradient field. Suppose the static field is uniform across the patient. If a rf pulse is applied at the Larmor frequency, &>0 = yB0, for a time long enough to create a 90° pulse, this will excite spins over a large volume of the patient. If, however, the z-gradient coil is used to apply an additional static field, Bz = Gzz, at the time of the rf pulse, the resonance condition will be met strictly only in the plane z = 0. Spins far from this plane are well off resonance and essentially will be unaffected by the rf pulse. Spins at z = 0 will be rotated by 90° just as if no gradient field were present. Spins close to, but not at, the plane z = 0 will be partially excited. To predict the exact behavior of the excitation, as a function of z, near the origin it is necessary to carry out a solution to the Bloch equations. It turns out that spins in the selected slice, but slightly off the center plane, will also be flipped through approximately 90° but will have phase differences with those spins at z = 0. This will reduce somewhat the signal generated by the slice. It can be shown that a more perfect slice profile will result if, instead of using a rf pulse of constant amplitude, the pulse amplitude is modulated by an appropriate envelope function. One useful modulation function is sin Ut/Ut. Here, U is an audio frequency that is high enough to permit sin Ut to go through a few cycles during the time the excitation pulse is being applied. Using this or slightly more complicated modulation functions, a rectangular slice profile can be approached. The stronger the gradient applied during excitation, the thinner the resultant slice will be. Typical gradient strengths are of the order of 1 G/cm or less. Typical durations for the excitation pulses are in the range from 1 to 3 msec. The slice thicknesses used in the early NMR scanners were relatively thick—on the order of 10-15 mm. Modern scanners are capable of producing 3- to 5-mm slice thicknesses routinely. The technique just described will excite a slice centered at z = 0. Additional audio-frequency modulation of the rf pulse can be used to move the location of the selected slice either up or down along the z axis.

We are using a coordinate system where the z axis points along the patient's body from the head toward the feet, the x axis is horizontal, and the y axis is vertical. The excitation method previously mentioned, which uses a z-gradient coil, will excite slices in the x-y plane; these slices are called axial planes. If, instead of a z gradient, a gradient in the x direction is applied during the rf pulse, the excited plane has a sagittal orientation. The x-z plane excited by a y-gradient field is called the coronal plane. By using a simultaneous combination of x, y, and z gradients various oblique planes may also be excited. Thus by a combination of electronically controlled rf pulses and gradient fields, planes of any orientation and location within the imaging volume may be excited as a first step in the imaging process.

B. Spin-Warp Technique

Several methods for converting NMR data into images have been suggested and demonstrated. The spin-warp technique has found the most wide spread clinical use and will now be described. The two-dimensional image to be formed consists of a large number of individual picture elements, called pixels. There are M rows of pixel elements in one direction and N columns in the other. For mathematical reasons M and N are both usually powers of two. For example, 128 x 256 and 256 x 256 are common pixel array sizes. The imaging process must yield a pixel brightness number for each of the M x N elements in the image. The basic ideas behind the use of gradient fields and Fourier transformation to create position-dependent frequency information are illustrated in Figs. 7 and 8.

The object to be imaged consists of the spins within a slice whose thickness t is determined by a selective excitation process. A desired field of view (FOV) is se-

FIGURE 7 Effect of a gradient field on the free induction decay. Water is located in two wells separated along the x axis. In (a) no gradient field is applied and a single damped exponential signal is seen. This is because the protons in both water samples have the same precession frequency. In (b) a gradient is applied along the x direction and a beat pattern is formed between the two frequencies that result. (Courtesy of GE Medical Systems.)

FIGURE 7 Effect of a gradient field on the free induction decay. Water is located in two wells separated along the x axis. In (a) no gradient field is applied and a single damped exponential signal is seen. This is because the protons in both water samples have the same precession frequency. In (b) a gradient is applied along the x direction and a beat pattern is formed between the two frequencies that result. (Courtesy of GE Medical Systems.)

lected within this slab and is divided into M x N volume elements called voxels. Normally the FOV is a square. Standard FOV sizes are 24 x 24 cm for head imaging and 40 x 40 cm for body imaging. For higher resolution studies of small anatomic regions, smaller FOVs (down to 8 x 8 cm) can be chosen, usually by varying the strengths of the applied gradients. The brightness to be assigned to each pixel in the image is proportional to the nuclear magnetization in the corresponding voxel. The size of the voxels in the x and y directions are given by Ax = FOV/N and Ay = FOV/M. The volume of a voxel is the product of the slice thickness t multiplied by (Ax Ay).

Figure 9 illustrates a pulse sequence that can produce the data for an axial image. The modulated rf 90° pulse and the simultaneous z gradient are used for the selective excitation of a plane centered at z = 0. Immediately after the rf excitation pulse is finished, a gradient pulse in the x direction is used to dephase the spins in the selected

FIGURE 8 Fourier transformation. In (a) the FID is a simple, damped-exponential function of time and its Fourier transform has a single peak at the corresponding frequency. In (b) the FID is a beat pattern consisting of two frequencies and has, correspondingly, two peaks in its transform. The width of a peak is inversely proportional to T2 that, in this case, is the same for both peaks. (Courtesy of GE Medical Systems.)

FIGURE 8 Fourier transformation. In (a) the FID is a simple, damped-exponential function of time and its Fourier transform has a single peak at the corresponding frequency. In (b) the FID is a beat pattern consisting of two frequencies and has, correspondingly, two peaks in its transform. The width of a peak is inversely proportional to T2 that, in this case, is the same for both peaks. (Courtesy of GE Medical Systems.)

FIGURE 9 Pulse and gradient timing diagram for spinwarp imaging. The top line shows the sequence of events that involve rf signals. The other three lines show the sequence of pulses on the three gradient coils. (Courtesy of Raven Press.)

slice. A 180° pulse is applied at the time t to refocus the spins, and thus a spin echo occurs at the time 2t. This maneuver permits the separatation in time of the excitation and receive periods and, therefore, the receiver electronics (which deal with a very low-level signals) are not forced to contend with any electronic ringing at the radio-frequency resulting from the very strong transmitter pulse. A pulse of y gradient is also used to give each line of spins at a fixed y position a different phase. The pulse is called the phase, encoding gradient and it is stepped in value each time the pulse sequence is repeated. This generates a different, y-dependent, phase shift for each cycle of the imaging process, and encodes, into the signal, information on the variation in spin density in the y direction. The receiver system is used to detect the voltage in the receiver coil during the sampling time Ts, which is centered on the maximum of the spin echo. A constant x gradient, called the readout gradient GR, is on during the sampling time. This causes the Larmor frequency to vary linearly in the x direction during the time that the signal is being received.

The signal received during Ts is composed of a narrow band of frequencies determined by the readout gradient. A filter is used to limit the detected signals to a bandwidth (BW). The voltage is sampled at N equal intervals during Ts. A criterion due to H. Nyquist states that the bandwidth, the sampling time, and N should be related by

For example, if BW = 32 KHz and N = 256, then Ts = 8 msec. The Nyquist criterion assures that if Eq. (10) is satisfied, all the information contained in the signal is also contained in the N digitized sample values. The BW is also related to the FOV by the relation

By combining Eqs. (10) and (11), the extent of the voxel in the x direction is given by

After a time TR measured from the beginning of the selective excitation pulse, the process is repeated for a total of M cycles, each of which uses a different value for the phase-encoding gradient. After this process is complete, a M x N array of digitally sampled data is available in the computer memory. This data can be converted by a two-dimensional Fourier transform technique into M x N pixel brightness numbers. These numbers can be displayed as an image, which can be viewed either on a cathode ray tube or as a hardcopy on film.

As an example of the voxel sizes used in MRI consider an image of the head using a 24-cm FOV, a 256 x 256 matrix size, and a 5-mm slice thickness. The value of 8x and 8y will be 240 mm/256 = 0.94 mm. The image will result from the 65,536 voxels in the object each with a volume of 4.7 mm3.

The phase-encoding process leads to Ay = FOV/M. If the sample contains any excited spins that lie outside the FOV in the phase-encoding direction, their signal will be added to the signals from the spins within the FOV and a form of image artifact called aliasing will result. The image is then a type of double exposure, with images of different parts of the anatomy superimposed on one another. If aliasing leads to an unacceptable level of confusion it can be dealt with by increasing M, while keeping the FOV constant (oversampling), and then displaying only the desired portion of the resulting image.

The time between the selective excitation pulse and the center of the sampling interval is called the echo time TE. Once the FOV and the slice thickness have been selected, the main imaging parameters that can still be varied are TE and TR. TE can be varied between roughly 20 and 200 msec. If TE is made long, a great deal of T2 relaxation can occur before the data is taken. In this case, only tissues with long values of T2 will give strong signals and will appear bright in the image. After each excitation the longitudinal magnetization will start to recover toward M0. The rate of this recovery is limited by Ti. If the repetition time TR is short, only those tissues with short values for T1 can become appreciably magnetized between excitations. Therefore, if it is desired to make a T1-weighted image, a relatively short TR is used and TE is made brief to prevent contrast based on T2 decay from developing. Conversely, a T2-weighted image can be created by using a long Tr (up to 2 sec between excitation pulses). This will permit all tissues to magnetize almost fully and eliminate contrast based on T1 differences. The use of a relatively long TE will allow differences in T2 decay rates to become manifest.

The total time required to complete a scan is M times TR, so that T2-weighted images generally take longer to acquire. Often it is desired to enhance the SNR by repeating the entire sequence one or more times and averaging the results of corresponding cycles. If there are n repetitions of the basic sequence, the total scan time increases to nMTR and the SNR is increased by -Jn. The total time to complete an individual scan usually ranges from about 1 to 15 min. Because it is usually necessary to make more than one series of images, the patient is normally in the magnet from 15 to 90 min to complete a diagnostic study.

Relaxation Audio Sounds Relaxation

Relaxation Audio Sounds Relaxation

This is an audio all about guiding you to relaxation. This is a Relaxation Audio Sounds with sounds called Relaxation.

Get My Free MP3 Audio

Post a comment