Essentials of MRI Safety. Donald W. McRobbie
can then be disconnected and removed from site. The reverse process, ramping down, can be used to reduce or remove the field when required, e.g. for a major hardware upgrade or after a non‐injurious ferromagnetic incident to remove the offending object.
The Nb‐Ti wires are 50–150 μm in diameter, embedded in a copper matrix. This provides additional mechanical strength – they are subject to significant magnetic forces – and provides a means for conducting excess heat and current in the event of a magnet failure or quench to prevent damage to the more delicate Nb‐Ti filaments. In the superconducting state, the copper matrix acts like an insulator, providing isolation between the Nb‐Ti strands.
Short larger‐bore magnets
A recent industry trend has been to reduce the length of the magnet, typical to around 1.6 m and to increase the diameter of the bore from 60 to 70 cm to afford better patient comfort and to accommodate larger patients. This has implications for safety as it affects the fringe field (see Fringe field spatial gradient, page 19).
Other magnets
Other configurations of MRI systems are also available, although less common. Resistive magnets producing fields up to 0.4 T are sometimes configured as open or C‐arm systems, affording better access to the patient and a less claustrophobic experience. Resistive magnets have one safety‐related advantage: the field can be routinely switched off.
Permanent magnets are used in low field niche scanners for extremity imaging or in “upright systems”. These employ various rare earth materials such as neodymium‐iron‐boron (Nd‐Fe‐B). Their magnetic field is always present.
Imaging gradients subsystem
Magnetic field gradients Gx, Gy, and Gz used to spatially select or encode the MR signal during acquisition are generated by three sets of gradient coils. The field generated is always along z. Gradient coils usually require water cooling as they have typically hundreds of amperes (A) of electricity pulsed through them. Specialist hybrid amplifiers and power supplies are used to generate these strong pulses. A consequence of gradient pulsing is the generation of acoustic noise (Chapter 7).
Gradient pulses usually have a trapezoidal waveform (Figure 1.16). The ability of the gradient system to switch rapidly, known as the slew rate (SR), is defined as the maximum amplitude divided by the rise time required to achieve that amplitude:
(1.4)
Figure 1.16 Trapezoidal gradient pulse.
Typical slew rates are 100−200 T m−1 s−1 (tesla per meter per second).
Example 1.2 Gradient performance
What is the (theoretical) maximum field produced by a 40 mTm−1 gradient system with a slew rate of 200 T m−1s−1? What is the minimum rise time?
Assuming that the gradient is linear over a 50 cm FOV, the maximum amplitude at the edge, 25 cm from the iso‐centre is
The minimum rise time is
Radiofrequency subsystem
The radiofrequency system comprises two subsystems: transmit and receive. RF transmit is more important for MR safety.
RF transmission
In most instances a body RF transmit coil is used (Figure 1.17). This typically has a “birdcage” design, operating in quadrature to produce a circularly‐polarized magnetic field B1. Some coils may operate in transmit and receive mode (T/R) denoted symbolically as in Figure 1.18. Examples are dedicated T/R head and knee coils. Tx coils usually have a cylindrical geometry, entirely encompassing the anatomical region to produce a uniform B1 so that everything in the FOV experiences the same flip angle.
Figure 1.17 Transmit 8‐rung ‘birdcage’ coil to produce a circularly polarised (rotating) B1+ field orthogonal to B0.
Figure 1.18 IEC 60601‐2‐33 [4] compliant coil labelling: left‐ transmit only; middle‐ transmit‐receive; right‐ receive only.
The coils operate in a resonant mode as tuned circuits, resulting in current amplification to achieve greater B1 at the Larmor frequency. They are driven by powerful RF amplifiers, rated at tens of kilowatts (kW). An important aspect of RF generation is impedance matching, usually to 50 Ω (ohms), to ensure the maximum power transfer from the amplifier to the coil. B1 is of the order of micro‐tesla (μT) peak amplitude.
In 3 T systems, operating at 128 MHz, the B1‐field in tissue is often quite non‐uniform. In this instance parallel transmit systems can help. These utilize multi‐element Tx coils powered by independent amplifiers capable of changing both the amplitude and phase (relative direction) of the RF pulses (Figure 1.19).
Figure 1.19 Parallel transmit: two (or more) independent RF power amplifiers drive elements of the transmit coil.
RF reception
The purpose of the RF receiver coils is to detect the tiny (micro‐volt) MR signals. A parallel tuned circuit is used (Figure 1.20) to magnify the voltage prior to pre‐amplification and further processing. The receive coil requires protection circuitry to prevent the large transmit pulses from coupling into the coil. A simple means of achieving this is to use crossed diodes and a detuning capacitor. During the large Tx pulses the diodes conduct and so the total capacitance becomes the sum of both capacitors and the circuit is off‐resonance. During signal detection the diodes do not conduct, and Cd is “invisible.”