Magnetic Resonance Imaging Physics

Issues of Concern

Gradient Coils

Spatial encoding is achieved with the help of the Gradient coils, which produce a secondary additional magnetic field superimposed over the primary magnetic field, thus varying the strength of the main field along the direction of the applied gradient field. They are arranged within the bore of the primary magnetic, i.e., X, Y, Z coils named after the axis of their orientation. Z coil produces axial images, Y produces coronal images, and x coil to produce the sagittal images. These coils are responsible for loud noises in MRI.[2]

Radiofrequency Coils

 Radiofrequency coils(RF coils) are used to transmit Radiofrequency Pulse ( RF) and receiving the signals from the area of interest. However, there are few transmit only and receive-only coils. There are different coils, i.e., head, shoulder, wrist, body, knee, etc. used for imaging the specific body parts.

Shim Coils

Good homogeneity is key for the acquisition of good quality MRI image. When the patient ( object) is placed within the main magnetic field, it creates the local susceptibility and reduces the homogeneity. Adjustments made to improve the homogeneity are called shimming, which could be passive or active. Passive shimming is achieved by placing sheets or metallic coins in a particular position. Active shimming is achieved by shim coils and which are activated and controlled by the host computer.

Basics of the MR Signal and Image

The Human Body is composed of 70 percent water(H2O). Hydrogen in the water and other molecules in body tissues consists of a single proton that carries a positive electric charge. The protons are constantly spinning and have their little magnetic fields. When there is no external applied magnetic field, they are randomly oriented. When an external Magnetic field B0 is applied, they align either parallel or antiparallel to each other. A slightly larger proportion of the protons align in a parallel direction ( low energy state ) than in the antiparallel direction ( high energy state). The net magnetic vector produced by the small number of extra protons lined parallel to the main magnetic field is the source of the MR signal and is used to produce MR images. The direction parallel to the main magnetic field is the longitudinal direction, which is called the Z direction. The plane perpendicular to this direction is called the transverse plane, i.e., X-Y Plane. The protons spin around a longitudinal axis like a precessing top at a rate called Larmor frequency, which is directly proportional to the magnetic field strength. For example, 1.5T produces a Larmor frequency of 64MHZ.[3]

RF pulse is applied for a few milliseconds disturbs or flips the protons at the same time and out of alignment from the main magnetic field. The amount of this rotation or flip depends on the strength and duration of the RF pulse. 90 RF pulse rotates the net magnetization vector to transverse plane .180 RF pulse rotates the net magnetization to –Z direction.

RF pulse can disturb the protons and transfer energy only when tuned to the precession frequency of the spinning electrons. This phenomenon is called resonance. When RF pulse is applied, some low energy protons flip to high energy state, decreasing the longitudinal magnetization, and protons begin to precess in phase. As a result, the net magnetization vector shifts to the transverse plane, i.e., the plane perpendicular to the main magnetic field. This generates an electric current, which is utilized to produce the MR image.[4]

T1 and T2 Relaxation 

When RF Pulse is switched off, the protons start to realign themselves back in a low energy state or equilibrium state, i.e., the main magnetic field's direction. This is called relaxation. Relaxation occurs in two different ways, i.e., T1 and T2 relaxation.

T1 Relaxation

Protons exchange energy with surroundings to return to their low energy state in the longitudinal axis, thus restoring the longitudinal magnetization called T1 relaxation ( spin-Lattice relaxation). T1 Time is the time taken for 63% of the longitudinal magnetization to be restored. Different molecules in our body have different T1 relaxation times, which would largely depend on how tightly the hydrogen atom is bound to the molecule. For example, water has long relaxation times and has a dark signal, whereas fat has a short T1 value and has a bright signal. T1 weighted images are useful for delineation of the anatomy.

T2 Relaxation 

Transverse magnetization starts to disappear, and protons fall out of phase in the X-Y plane, and this process is called transverse magnetization. T2 time is the time where only 37% of the original transverse magnetization is present. There are many causes for the loss of coherence in protons, i.e., spin-spin interaction, magnetic field inhomogeneity, Magnetic susceptibility, and chemical shift artifact. Dephasing that occurs due to all these factors is called T2* relaxation.

Firstly, inhomogeneity within the local tissue causes protons' spin to be influenced by the small magnetic fields by spin from a neighboring molecule. This internal inhomogeneity influencing the spins of the neighboring protons is called spin-spin relaxation or T2 relaxation. T2 decay is very characteristic of a specific tissue. For example, free water has relatively apart molecules hence have long T2 times compared to a large macromolecule.[4]

T1 and T2 Weighted Images

T1 and T2 relaxation times largely govern tissue contrast. Time to Echo (TE) is the time between the peak of 90 degrees RF pulse and the peak of the echo formed. TR is the time taken to run a pulse sequence at one time. TE and TR control the number of weighing effects in the MR image. An image produced predominantly due to difference in T1 relaxation times is called T1 Weighted images. T1 weighted images are produced by mainly manipulating the time between TE and TR. When the T1 effects are minimized by manipulation of TE and TR, T2 weighted images are produced. T2 weighted images have short TE and short TR, whereas the T1 weighted images have short TE and long TR. Interestingly, long TR and short TE decreases both T1 and T2 effects and produces protein density images.[5]

Free Induction Decay

Once the RF pulse is switched off, protons continue to precess and follow a spiral path with changing magnitude and direction. The highest magnitude is observed immediately after the RF pulse is switched off,  and the FID signal obtained demonstrates a sine wave pattern.

Basic Sequences

Spin Echo

This sequence produces T1 weighted, T2 weighted, Proton density images by manipulating the TE and TR. Typical values of TE and TR for T1 weighting (at 1.5 T) are TE = 20 msec and TR = 500 msec and  the typical values for T2 weighting are TE = 80 msec and TR = 2,000 msec. 1800 Pulse is applied at half TE time, and echo is centered at TE. Multiple Spin echo sequence uses multiple 1800 Pulses to generate multiple echoes, i.e., TR is usually 2000msec, but with TE1 at 20msec and TE2 at 80 msec, both T2 and protein density images can be produced.[6]

Turbo Spin Echo

Turbo spin-echo differs from spin echo in that it applies multiple 180 pulses during one TR. The number of these refocusing pulses is determined by echo train length (ETL). Hence pulse sequence is much faster, i.e., ETL of 16 would make the sequence 16 times faster.[7]

Gradient Echo

Gradient echo sequences replace the 180 refocusing pulse with the magnetic field gradient pulse. After the initial RF pulse, a gradient pulse is applied, and this causes the protons to rapidly dephase along the gradient direction and rapid decline of the FID signal. This loss of phase coherence is reversed by adding a second gradient pulse of equal magnitude but opposite direction to the first. This causes the proton to move back in phase and results in Gradient echo (GRE images). The difference between the gradient echo and spin echo is short TE as only one RF pulse is used, and also the flip angle is typically 5- 40. This sequence is particularly useful for detecting hemorrhage and has been particularly useful in detecting various pathologies like AVM, superficial siderosis, and cavernomas.[8]

Inversion Recovery

This sequence utilizes an initial 180 pulse instead of 90 pulses in the spin-echo sequence. This causes the initial inversion of the longitudinal magnetization, which then slowly begins to grow back in the main magnetic field direction. Different tissues grow back at different rates. When the tissue of interest crosses the zero axes, a 90 pulse will cause all signals from other tissues to rotate to the transverse plane. Note that the tissue of interest will not contribute to any brightness of the images as it is at zero axes. This sequence is widely used to obtain fat-suppressed images.[9]

Intravenous Contrast

Gadolinium, a paramagnetic substance, is used as an intravenous contrast agent. It has a strong paramagnetic property with numerous unpaired electrons resulting in shortening both T1 and T2 values in tissues. It is seen as the bright signal in T 1 weighted images and is widely used in clinical practice. The incidence of allergy to the Gadolinium is very rare compared to the iodine-based CT contrast agents (0.03%). Although rare, in patients with severe renal impairment, it can cause nephrogenic systemic fibrosis(NSF).[10]

MRI Safety

 Magnetic fields generated by the MRI machine is very strong. For example, 1.5T can generate a magnetic field around 21000 times that of the earth’s natural field. This can cause metallic objects to move suddenly and can cause injuries. Hence, it is important to remove all metallic belongings like hearing aids, belts, and jewelry before the scan, and also pagers, cameras, and cell phones should be turned off in the MRI examination room. It is also important to let the technician know about any internal implants like aneurysm clips, a pacemaker, or any metallic foreign body to undertake appropriate screening. A precheck questionnaire helps to ensure safety.[11]

In the case of life-threatening situations like fire, a process called quenching is undertaken. This releases the compressed liquid helium through the vent, and the machine loses superconducting magnet properties.

MRI and Pregnancy

MRI utilizes nonionizing radiation is relatively safe in pregnancy. However, there is a theoretical risk that excessive heating effects caused by MR gradient changes may cause teratogenic effects like reduction in the Crown-rump length (CRL). Hence, it is best avoided in the first trimester where possible. The use of a Gadolinium-based contrast agent is avoided wherever possible and should be used when the benefits outweigh the risk.[12]

Clinical Significance

Knowledge of MRI physics is essential to the appropriate interpretation of MRI images. Due to excess demand and also wide availability of MRI, MR studies are frequently reviewed even by clinicians. Hence a sound knowledge of MR physics is essential for both radiologists and clinicians for adequate interpretation of MRI images.