That has now changed, largely due to the introduction of the picture archiving and communication system. More frequently, clinicians review MRI before seeking specialist radiological opinion. However, knowledge of the basic physical principles underlying MRI acquisition is fundamental to image interpretation. The MR system comprises two main groups of equipment. The first is the control centre, which is positioned where the operator sits. Its associated electronics and power amplifiers are usually situated in an adjacent room and connect to the second equipment group.
This second group of equipment is housed within the machine in which the patient lies. It contains the parts of the MR system that generate and receive the MR signal and include a set of main magnet coils, three gradient coils, shim coils and an integral radiofrequency RF transmitter coil 1 figure 1. Due to the necessary use of RF electromagnetic waves or radio waves see below , the room that contains this second set of equipment needs to keep potential sources of electromagnetic noise out and its own RF in.
This is achieved by enclosing the magnet and its associated coils within a special, copper-lined examination room, forming what the Physics community calls a Faraday shield. Schematic demonstrating the relative positions of the different magnet coils comprising the MR machine. The patient is positioned within the bore of the machine and is surrounded by coils that lie concentric to each other and in the following order from furthest to closest to the patient: For neuroimaging, a further RF coil is placed around the patient's head to improve signal to noise ratio.
Recall the principles of the Maxwell equations that indicate that when an electric current flows through a wire, a magnetic field is induced around the wire. Resistance to the flow of the electric current can be reduced to negligible levels if a special metal conductor is cooled substantially.
In this situation, lower resistance allows the use of high electric currents to produce high-strength magnetic fields, with little heat disposition. This principle is employed in the generation of superconducting magnets: The main magnet coils generate a strong, constant magnetic field B 0 to which the patient is exposed.
The strength of the magnetic field is measured in units of Tesla, T. Currently, most clinical MR systems are superconducting and operate at 1. Field strengths reaching 9. The MR system uses a set coordinates to define the direction of the magnetic field. Gradient coils representing the three orthogonal directions x, y and z lie concentric to each other within the main magnet figure 1. They are not supercooled and operate relatively close to room temperature. Each gradient coil is capable of generating a magnetic field in the same direction as B 0 , but with a strength that changes with position along the x, y or z directions, depending on which gradient coil is used.
The magnetic field generated by the gradient coils is superimposed on top of B 0 so that the main magnetic field strength varies along the direction of the applied gradient field figure 2. Image shown depicts the generation of a gradient in B 0 in the Z direction. For a standard clinical MR system, this is accomplished using two coils in which the current flowing through them runs in opposite directions to each other the so-called Maxwell pair type.
The magnetic field at the centre of one coil adds to the B 0 field while the magnetic field at the centre of the other subtracts from B 0 , thus creating a gradient in the B 0 field. RF coils, so named because the frequency of electromagnetic energy generated by them lies within the megahertz range, are mounted inside the gradient coils and lie concentric to them and to each other.
Some RF coils perform the dual role of transmission and reception of RF energy whereas others transmit or receive only. For neuroimaging, a separate RF receiver coil that is tailored to maximise the signal from the brain is usually applied around the patient's head to detect the emitted MR signals.
The RF field is also referred to as the B 1 field. When switched on, the B 1 field combines with B 0 to generate MR signals that are spatially localised and encoded by the gradient magnetic fields to create an MRI. Localisation of the MR signal requires good homogeneity within the local magnetic field. In other words, the more uniform the magnetic field the better. However, placement of an object including a patient within the main B 0 field creates local susceptibility effects and reduces homogeneity. Shimming refers to adjustments made to the magnet to improve its homogeneity.
Shimming can be passive or active. Passive shimming is achieved during magnet installation by placing sheets or little coins of metal at certain locations at the edge of the magnet bore close to where the RF and gradient coils lie. Active shimming provides additional field correction around an object of interest through the use of shim coils, which are activated by electric currents controlled by the host computer, under the guidance of the scanner application software and the operator.
Homogeneity of and hence variation in B 0 is quoted in parts per million ppm , that is, a fraction, of the static magnetic field over a specified spherical volume. The primary origin of the MR signal used to generate almost all clinical images comes from hydrogen nuclei. Hydrogen nuclei consist of a single proton that carries a positive electrical charge. The proton is constantly spinning and so the positive charge spins around with it. Recall that a moving electrical charge is called a current and that an electrical current generates a magnetic field.
Thus, protons have their own magnetic fields and behave like little bar magnets figure 3. Protons possess a positive charge and are constantly spinning around their own axes. This generates a magnetic field making protons similar to bar magnets. This figure is only reproduced in colour in the online version.
The magnetic field for each proton is known as a magnetic moment. Magnetic moments are normally randomly orientated. However, when an external magnetic field B 0 is applied they align either with parallel or against antiparallel the external field. The preferred state of alignment is the one that requires the least energy: Accordingly more protons align with B 0 than against it. The difference in the number of protons aligning parallel and antiparallel to B 0 is typically very small but ultimately depends on the strength of B 0 as well as the temperature of the sample.
When put in an external static magnetic field, the overall effect on a group of protons that individually are aligned either parallel or antiparallel to B 0 means that the group of spins classically move in a particular way called precession. Precession can be likened to the movement of a spinning top.
When spun, the top wobbles but does not fall over and the axes of the top circles form a cone shape figure 4. When exposed to an external magnetic field, protons precess. The movement of precession can be likened to the wobbling motion seen when a spinning top is spun. The handle of the spinning top follows a circular path. Its value for the proton is The Larmor equation indicates that precession frequency is proportional to the strength of the magnetic field.
Protons precessing parallel to B 0 begin to cancel each other out in all directions bar one: The result is a sum magnetic field or sum magnetisation, often given the symbol M, with the value M 0. It is characteristically shown as a vector. As this sum magnetisation parallels the external magnetic field it is also referred to as longitudinal magnetisation figure 5 B. For simplicity protons are now shown as vectors. A The magnetic moments of protons precessing in the external magnetic field begin to cancel each other out. Protons precessing parallel to B 0 also begin to cancel each other out.
General MRI Terms
This occurs in all directions bar one: The result is a sum magnetic field that is typically depicted as a vector B. The patient essentially becomes a magnet with a magnetic vector aligned with B 0. The magnetic force of the patient as it stands cannot be measured as it is in the same direction as the external field.
What is required is a magnetisation that lies at an angle to B 0. With the patient in the magnet and possessing longitudinal magnetisation, RF pulses are switched on and off. The purpose of the RF pulse is to disturb the protons so that they fall out of alignment with B 0. This disturbance occurs through the transference of energy from the RF pulse to the protons. This can only occur when the RF pulse has the same frequency as the precessional frequency of the protons, a phenomenon called resonance; hence the term magnetic resonance imaging.
The activation of an RF pulse has two main effects on the protons. First, some protons gain energy and move to the higher energy state of being antiparallel to B 0. Second, the RF pulse causes the protons to move in phase ie, in the same direction, at the same time with each other rather than in random directions.
The result is transverse magnetisation in which a new magnetisation vector is created in the x—y plane and moves in line with the precessing protons at the Larmor frequency. The transverse magnetisation vector is a moving magnetic field rotating at the Larmor frequency and, as such, if a conductive receiver coil is placed in proximity, an alternating voltage will be induced across it. This in turn generates an electrical current, which can be picked up like an antenna would pick up radio waves forming an MR signal. As soon as the RF pulse is switched off the protons start to fall out of phase with each other and also return to a lower energy state, that is, the protons relax.
Relaxation occurs in two different ways. A Protons aligned with B 0 produce a sum vector with longitudinal magnetisation. B When an RF pulse is switched on longitudinal magnetisation decreases and transverse magnetisation propagates. Immediately following an RF pulse, protons precess in phase in the transverse plane, depicted by a single vector arrow in the lower circle.
During this process, the whole system continues precessing and so the sum vector takes a spiralling motion D. Recovery of longitudinal magnetisation is termed T1 relaxation and loss of transverse magnetisation is called T2 relaxation. T1 relaxation is the process whereby protons exchange energy with their surroundings to return to their lower energy state and in doing cause the restoration of longitudinal magnetisation. The rate at which this occurs is dependent on the tumbling rate of the molecule in which the proton resides.
Tumbling rate describes the rate of molecular motion. As molecules tumble they generate a fluctuating magnetic field to which protons in neighbouring molecules are subjected. Energy exchange and therefore T1 relaxation is more favourable when this fluctuating magnetic field is close to the Larmor frequency. Different molecules have different tumbling rates and as a result they also differ in their efficiency at T1 relaxation. Similarly, hydrogen protons bound to large macromolecules eg, membrane lipids tumble very slowly and also demonstrate low efficiency at T1 relaxation.
Conversely, when water is partially bound to proteins for example its tumbling rate can be slowed to a rate more in line with the Larmor frequency. This is because the carbon bonds at the ends of the fatty acids have frequencies near the Larmor frequency, allowing effective energy transfer. As T1 relaxation requires an exchange of energy between protons and their surroundings it is also termed spin—lattice relaxation. The term lattice is a throwback to early nuclear MR studies in solids, in which the external environment was literally a crystalline lattice of molecules.
As not all protons return to their original energy state at the same time, T1 relaxation is more of a continuous process. Plotting the recovery of longitudinal magnetisation over time after the RF pulse is switched off produces an exponential curve, called the T1 curve. It is difficult to exactly pinpoint the end of longitudinal relaxation and so T1 and similarly for T2 is not defined as the time of completion of longitudinal relaxation.
Rather, T1 is a time constant that is used to describe how fast the process of T1 relaxation takes. Plotting the recovery of longitudinal magnetisation over time following the switching off of a radiofrequency RF pulse results in a T1 curve. This results in a temporary gain in signal intensity at time echo time TE termed spin echo. Each subsequent echo will be of lower intensity due to T2 effects. A curve connecting the spin echo intensities is the T2 curve. Transverse relaxation describes the process whereby protons fall out of phase in the x—y plane and transverse magnetisation decreases and disappears.
There are two causes for this loss of phase coherence. The first, T2 relaxation results from slowly fluctuating magnetic field variations inhomogeneity within the local tissue. That is, the magnetic spin of protons is influenced by small magnetic fields from neighbouring nuclei. This results in the random fluctuation of the Larmor frequency of individual protons causing an exchange of energy between proton spins, which leads to loss of phase coherence across a population of protons.
The second cause of loss of phase coherence is due to inhomogeneity within B 0. Magnetic field variations result in slightly different Larmor frequencies for protons at different locations within the field. Unlike the random process of T2 relaxation, this de-phasing is caused by a constant and is potentially reversible. Spin—spin interaction governs the speed of T2 relaxation and hence influences the T2 values for different tissues. In contrast to T1 relaxation, where energy transfer from the spin system must occur, T2 relaxation may proceed with or without overall energy loss.
For human tissue transverse relaxation is typically a much faster process than longitudinal relaxation; hence, T2 values are always less than or equal to T1. When an RF pulse is switched off, T1 and T2 relaxation occur simultaneously and independently. The protons continue to precess and the sum magnetisation vector follows a spiralling path whereby its direction and magnitude are constantly changing. Hence, an electrical signal is generated in a suitable receiver coil.
It has its greatest magnitude immediately after the RF pulse is switched off and then decreases as both relaxation processes occur. It also has a constant frequency resonant frequency and consequently the FID signal takes the form of a sine wave with a rapidly decaying envelope figure 8. Its amplitude along with signal intensity then decreases as the protons begin to lose phase coherence. The resultant decay signal is termed free induction decay. FID is subjected to further disruption de-phasing by the magnetic field gradients that are used to localise and encode the MR signal.
Instead, it is common practice to generate and measure the MR signal in the form of an echo: Echoes can be appreciated by considering how T1- and T2-weighted images are formed. Contrast between tissues allows adjacent structures to be differentiated from one another. Contrast is determined by signal intensities, which in turn are governed at least partly by the T1 and T2 relaxation times of tissues within an image.
An image in which the difference in signal intensity between tissues is predominantly due to differences in tissue T1 relaxation time is called a T1-weighted image. T1-weighted images are generated predominantly by manipulating the time between two RF excitation pulses, the so-called repetition time TR figure 9. Repetition time TR and T1-weighting.
MRI: The Basics - Ray H. Hashemi, William G. Bradley, Christopher J. Lisanti - Google Книги
Consider the following example in which two tissues, fat and fluid are being imaged where fat has shorter transverse and longitudinal relaxation times than fluid. As MR signal sampling occurs after the switching off of the second RF pulse, the signal intensity will be determined by the amount of transverse magnetisation at that point.
In this situation, the application of a second RF pulse with subsequent MR signal sampling would result in no discernible difference in MR signal between the tissues as the transverse magnetisation values for the tissues are essentially equal A ie, no signal contrast. The tissues will possess different transverse magnetisation values B and, thus, will generate different signal intensities and allow greater contrast between tissues C.
The preceding example was simplified for ease of understanding. It should be appreciated that many parameters influence signal intensity and hence tissue contrast but in the example used T1 had the greatest influence. Contrast in images obtained at long TR will not be influenced by T1 but instead may be influenced by differences in the T2 or proton density of the tissues in question. Factors that influence MR signal intensity are listed in the box. Box 1 Factors that influence MR signal intensity Proton density.
De-phasing caused by T2 relaxation is a random, irreversible process whereas the de-phasing caused by magnetic field inhomogeneity is potentially reversible. Local field inhomogeneity remains and protons with a slightly faster Larmor frequency begin to catch up with slower protons. Eventually the protons come back into phase, which results in an increase in the amplitude of the MR signal. Maximum signal amplitude is reached at the echo time TE. Over time, the amplitude from each SE will decrease due to T2 effects.
A curve connecting the SE intensities is called the T2 curve.
- The Journal of a Mission to the Interior of Africa, in the Year 1805;
- Dead Heat: Global Justice and Global Warming (Open Media Series);
- Lessing - German Playwright.
- Camp Sites: Sex, Politics, and Academic Style in Postwar America (Post*45)!
- Understanding MRI: basic MR physics for physicians | Postgraduate Medical Journal;
At time TE the protons regain phase, resulting in a stronger net transversal magnetisation and thus a stronger signal. This signal re-emergence is termed spin echo. Tissues have different T2 values.
MRI Technique
Brain for example has a shorter T2 than cerebrospinal fluid. It is one of the parameters whose value can be chosen by the operator of the MR machine in order to influence the signal intensities hence contrast between tissues. A much stronger signal is received when short TE as opposed to long TE is employed. Echo time TE and T2-weighting.
T2 curves for two different tissues are shown. Tissue B has a shorter T2 relaxation time than tissue A. The difference in signal intensity between the tissues is more discernible at long TE than at short TE. A heavily T2-weighted image could be obtained at longer TE but loss of MR signal would impact on the signal to noise ratio making images potentially subdiagnostic. Values for both of these parameters are purposely chosen by the operator in order to influence the tissue weighting of the image.
A different type of image is produced at long TR and short TE. When this occurs, the signal is predominantly influenced by the proton density of the tissues figure 12 A—C. Examples of typical clinical MRI. Note that fat around scalp and neck is bright and cerebrospinal fluid CSF is dark. B Axial proton density image.
- MRI technique - Startradiology.
- Glorious Christianity (Studies in the Book of Acts);
- 10 Commandments of Christian Conflict Resolution (Self Help Book 12).
Note minimal contrast between grey matter and CSF. C Axial T2-weighted image. Note that CSF is bright. D Same image as A but following gadolinium administration. Note the dural venous sinuses appear bright due to T1 shortening. E Axial fluid attenuated inversion recovery image. Note suppression of the CSF signal. Same patient as in C. Note the increase conspicuity of multiple dark foci due to magnetic susceptibility of haemosiderin deposition in a patient with amyloid angiopathy.
It should be appreciated that the choice of imaging parameters for all MRI sequences can influence the sensitivity of the test for the pathology in question. For example, the sensitivity for the detection of lesions in multiple sclerosis is dependent on TE lesion visibility decreases with increasing TE. This limits the number of patients who can be scanned in a session and also risks movement artefact by a restless patient.
However, they do produce different contrasts. Magnetic field gradients produce a change in field strength and thus a change in Larmor frequency along a particular direction. Application of a gradient pulse after an initial RF pulse causes protons to rapidly de-phase along the direction of the gradient resulting in rapid decline in the FID signal.
This loss of phase coherence can be reversed by applying a second magnetic field gradient with a slope of equal amplitude but in opposite direction to the first. As a result, protons move back into phase and return a signal called GRE. TE is the time taken between the beginning of FID ie, generation of transverse magnetisation following the initial RF pulse to the point at which the GRE reaches its maximum amplitude figure An excitatory radiofrequency RF pulse causes transverse magnetisation and initiation of a free induction decay signal. This signal rapidly de-phases following the application of a magnetic field gradient.
Application of a second magnetic field gradient with a slope of equal amplitude but in opposite direction to the first causes some rephasing. The signal increases again at time TE to a maximal signal termed a gradient echo. Several differences exist between the two techniques. As a consequence, longitudinal magnetisation is not completely abolished with GRE and will provide reasonable signal even at very short TR.
The price of short acquisition time in GRE, however, is greater signal loss in the presence of magnetic susceptibility effects. This is discussed further in the section on magnetic susceptibility. The MR signal is localised in three dimensions using three separate magnetic field gradients termed 1 slice-selection gradient, 2 phase-encoding gradient G P and 3 frequency-encoding gradient G F.
Slice localisation is achieved by using gradient coils to generate a gradient field orientated along a chosen axis. This gradient field alters the strength of B 0 in the chosen direction, so that protons within the gradient field have different Larmor frequencies figure 14 A. A slice-selecting gradient G S is applied at the same time as the excitatory radiofrequency RF pulse. In this example, the brain is being imaged and G S is applied along the z-axis, parallel with B 0. This means that different cross sections of the brain experience magnetic fields of differing strength.
Accordingly protons will precess at different Larmor frequencies depending on their position along the gradient. Selecting an RF pulse or range of RF pulse frequencies that matches the Larmor frequency of the protons will determine the slice location. Protons are in phase after the RF pulse is applied.
Applying a phase-encoding gradient along the y-axis causes the protons to increase their speed of precession relative to the strength of the magnetic field to which they are exposed. In this example, speed increases from top to bottom. This is depicted on the right of the schematic, which shows rows of protons with different precession speeds. When the gradient is switched off, all protons are once again exposed to the same magnetic field and as such they have the same frequency; only now, they are out of phase.
It may help to think of the protons giving of the same signal frequency but at different times. A further magnetic field frequency-encoding gradient is applied to help differentiate the signal from different protons by way of differing frequencies. In this example, the protons from the bottom row in step 2 above have been exposed to a gradient applied along the x-axis.
While the protons are in phase, they now have different frequencies, allowing their differentiation in the third x plane. It can also detect demyelinating disease, and has no beam-hardening artifacts such as can be seen with CT. Imaging is also performed without any ionizing radiation. MRI is based on the magnetization properties of atomic nuclei. A powerful, uniform, external magnetic field is employed to align the protons that are normally randomly oriented within the water nuclei of the tissue being examined. This alignment or magnetization is next perturbed or disrupted by introduction of an external Radio Frequency RF energy.
The nuclei return to their resting alignment through various relaxation processes and in so doing emit RF energy. After a certain period following the initial RF, the emitted signals are measured. Fourier transformation is used to convert the frequency information contained in the signal from each location in the imaged plane to corresponding intensity levels, which are then displayed as shades of gray in a matrix arrangement of pixels.
Repetition Time TR is the amount of time between successive pulse sequences applied to the same slice. Tissue can be characterized by two different relaxation times — T1 and T2. T1 longitudinal relaxation time is the time constant which determines the rate at which excited protons return to equilibrium. It is a measure of the time taken for spinning protons to realign with the external magnetic field.
T2 transverse relaxation time is the time constant which determines the rate at which excited protons reach equilibrium or go out of phase with each other. It is a measure of the time taken for spinning protons to lose phase coherence among the nuclei spinning perpendicular to the main field. T1-weighted images are produced by using short TE and TR times. The contrast and brightness of the image are predominately determined by T1 properties of tissue.
In these images, the contrast and brightness are predominately determined by the T2 properties of tissue. CSF is dark on T1-weighted imaging and bright on T2-weighted imaging. By doing so, abnormalities remain bright but normal CSF fluid is attenuated and made dark. This sequence is very sensitive to pathology and makes the differentiation between CSF and an abnormality much easier. T1-weighted imaging can also be performed while infusing Gadolinium Gad.