Author: Mohamed Abdelghani
Abstract
Diagnosis is one of the most important steps during treatment. However, it may be difficult and not so accurate. Accuracy may be achieved by imaging tools, yet not all of them are safe. At this moment, magnetic resonance imaging appears to be one of the most effective imaging tools for not only diagnosis but also for surgical planning and even treatment. MRI is a complex yet effective technique. MRI is a unique technique because of its high spatial resolution, unique differentiation, and depth penetration. Globally, there are about 36,000 MRI machines. Since Paul Lauterbur and Peter Mansfield published the technique called MRI, it is now a critical tool. MRI is used in a variety of fields, such as clinical studies and diagnosis, research studies, and brain development. This paper will show what MRI is, focusing on its mechanism from the beginning till image acquisition. Also, it will mention some of the advantages and disadvantages of MRI.
Keywords: Magnetic Resonance Imaging, T1 Relaxation, T2 Relaxation.
Introduction
Many years ago, diagnosis was not as we know it today. Imaging tools revolutionized diagnosis without a single cut. The development of MRI for use in medical investigation has provided a huge forward leap in the field of diagnosis. MRI offers avoidance of exposure to potentially dangerous ionizing radiation; this specific aspect makes MRI a unique imaging tool (Grover et al., 2015).
MRI can evaluate organs like the brain, liver, kidneys, lungs, and spleen (Manikkavasakar, 2014; Palas et al., 2013; Kumar et al., 2016; Selby et al., 2018). The diagnostic importance of MRI is reflected by the growing number of MRI scanner installations (Moser et al., 2008).
The paper will discuss the principles of MRI with its advantages and disadvantages while focusing on the mechanism of how MRI works.
Introduction to MRI
Magnetic resonance imaging (MRI) is a non-invasive imaging tool; it doesn’t break the skin or harm the body (Nelson, 2008; Su et al., 2020). MRI can visualize organs and soft tissues in great detail. Unlike many imaging techniques like X-rays and computed tomography (CT) scans, MRI has no ionizing radiation. Thus, at most field strengths (generally below 7 Tesla), an MRI scan is considered safe for nearly every age group. MRI is used in many fields, from clinical studies to studies of brain development (Nelson, 2008; Jensen et al., 2015).
The first nuclear magnetic resonance (NMR) signals from a living animal were acquired from an anesthetized rat in 1968. Firstly, the capability to differentiate tumors from normal tissue by NMR was reported by the American physician Raymond Damadian, the inventor of NMR, in 1971. MRI was developed by the American chemist Paul Lauterbur based on the encoding of spatial information of NMR signals with magnetic field gradients. The first cross-sectional image of a living mouse was published in 1974 by Lauterbur. The echo-planar imaging (EPI) technique was developed by the English physicist Peter Mansfield (Jahng et al., 2020). The first MRI body scan of a human was performed by Damadian in 1977 (Vassiliou et al., 2018), as shown in Image 1.
Image 1: The first MRI image of a human body, obtained in 1977 by Damadian.
For their efforts, Paul Lauterbur and Peter Mansfield were awarded the Nobel Prize in Physiology and Medicine in 2003. As a summary, the timeline of MRI development and a summary of the major contributions are shown below. Where BOLD denotes blood oxygen level-dependent; UHF denotes ultrahigh field; NIH denotes National Institutes of Health; PET denotes positron emission tomography; and LINAC denotes linear accelerators (Jahng et al., 2020).
The timeline of MRI development and a summary of the major contributions.
Components of MRI
An MRI scanner consists of several elements including a magnet with a bore creating an external magnetic field; permanent magnets (with lower induction of magnetic field) or a superconducting magnet cooled down by liquid helium, which is more common; the elements are as below:
• gradient coils—modifying external magnetic field;
• radio frequency transmitters—emitting and receiving signal;
• matrix—registering signals;
• patient table;
• computer system—controlling MRI unit;
• technical console—programming of examination protocol and control throughout the examination;
• if contrast-enhanced examinations are required, a contrast agent is used, a liquid (usually gadolinium) that is injected into the body.
A schematic drawing of the MRI scanner is shown in figure 1 (Delantoni & Orhan, 2022).
Figure 1: A schematic drawing of the MRI scanner.
Mechanism of MRI
MRI employs radio frequency (RF) radiation in the presence of controlled magnetic fields. Generally, there are two ways of explaining the fundamentals of MRI: classical theory and quantum theory. The principles of MRI rely on the spinning motion of specific nuclei present in tissues. The number of spin values depends on the atomic and mass numbers. If the number of neutrons and the number of protons are both even, the nucleus has no spin. If the number of neutrons plus the number of protons is odd, then the nucleus has a half-integer spin. In nuclei with an even mass number caused by an even number of protons and neutrons, half of the spin of the nucleon is in one direction and half in the other; the forces of rotation cancel out. The nucleus itself has no net spin, or angular momentum. Only nuclei with an odd mass number or atomic weight are used. These are known as MR active nuclei (Westbrook & Talbot, 2018).
MR-active nuclei have the tendency to align their axis of rotation with an applied magnetic field. The reason for that is that they have angular momentum, or spin. Another reason is that they possess an electric charge, as they contain positively charged protons. The law of electromagnetic induction, the one that was determined by Michael Faraday in 1833, refers to the connection between electric and magnetic fields and motion. Faraday’s law determines that a moving electric field produces a magnetic field and vice versa. MR-active nuclei have a net electrical charge and are spinning and, therefore, automatically acquire a magnetic field. In classical theory, this magnetic field is denoted by a magnetic moment. The magnetic moment of each nucleus has vector properties, i.e., it has size (or magnitude) and direction. The total magnetic moment of the nucleus is the vector sum of all the magnetic moments of protons in the nucleus (Westbrook & Talbot, 2018).
A hydrogen isotope called protium is the most used MR-active nucleus in MRI. It has a mass and atomic number of 1. It is used because hydrogen is very abundant in the human body and also because the solitary proton gives it a relatively large magnetic moment (Westbrook & Talbot, 2018; Currie et al., 2012). These characteristics mean that the maximum amount of available magnetization in the body is utilized. Faraday’s law states that a magnetic field is created by a charged moving particle (that creates an electric field). The protium nucleus contains one positively charged proton that spins, i.e., it moves. Thus, the nucleus has a magnetic field induced around it and acts as a small magnet. The magnet of each hydrogen nucleus has a north and a south pole of equal strength. The north or south axis of each nucleus is represented by a magnetic moment and is used in classical theory. As shown in figure 2, the magnetic moment is represented by an arrow (Westbrook & Talbot, 2018).
Figure 2: The magnetic moment is shown by an arrow. The arrow’s length represents the magnitude of the magnetic moment or the strength of the magnetic field surrounding the nucleus. The direction of the arrow denotes the direction of alignment of the magnetic moment.
In the absence of a magnetic field, the magnetic moments of hydrogen nuclei (protons) are randomly orientated and produce no overall magnetic effect. Nevertheless, when placed in a strong static external magnetic field, termed B0, their magnetic moments orientate with it. This is called alignment (Westbrook & Talbot, 2018; Mastrogiacomo et al., 2019). In figure 3, a comparison between random alignment with no external field applied and alignment after applying an external magnetic field B0 is shown. The alignment with the B0 happens on the z-axis (Westbrook & Talbot, 2018).
Figure 3: A comparison between random alignment with no external field applied and alignment after applying an external magnetic field B0.
This can result in two ways: either in the direction of the field, spin up, or anti-parallel (opposite) to the field, spin down. These orientations correspond to lower energy states and higher energy states of the dipole respectively. As it is easier for the nuclei to be parallel with the magnetic field, this is considered a low-energy state. Energy can be supplied or recovered in the form of electromagnetic energy in the RF portion of the electromagnetic spectrum, and this transition from one energy level to another is called resonance (Katti et al., 2011).
When B0 is applied, the North and South poles do not align exactly with the direction of the magnetic field. The axes of spinning protons oscillate with a slight tilt from a position that is parallel with the flux of the external magnet. This tilting is called precession. The precession of protons is shown in figure 4.
Figure 4: The precession of protons.
The rate or frequency of precession is called the Larmor frequency. The Larmor frequency of hydrogen is 42.58 MHz in a magnetic field of 1 Tesla. The magnetic field strengths used for MRI range from 0.1 to 4.0 T. The Larmor equation is shown in Equation 1.
Equation 1: The Larmor equation
Here, ω is the Larmor frequency; γ is the gyromagnetic ratio; B0 is the applied magnetic field. Next, the RF pulse is applied, and the energy that is in the form of all electromagnetic waves from an RF antenna coil is directed to tissue that contains hydrogen nuclei. The nuclei with the same frequency absorb the energy and shift to the xy-plane (Katti et al., 2011).
As mentioned, if the pulse is sufficient, it will rotate the net tissue magnetization net into a transverse plane. These planes are perpendicular to the longitudinal alignment (z-axis). This causes all the protons to precess in phase; this is referred to as a 90 degrees RF pulse or a flip angle of 90 degrees. The RF signal is induced in a receiver coil. In figure 5, the scenario for the net magnetic moment before applying the field until it tips to another axis is shown (Delantoni & Orhan, 2022; Mastrogiacomo et al., 2019).
Figure 5: The scenario for protons before applying the field until they tip to another axis.
After it tips, it keeps precession, but now precession is on the xy plane. And then, the signal is detected by the coil, as shown in figure 6 (Möllenhoff et al., 2012).
Figure 6: The precession of protons in the xy-plane and receiving the signal by the coil.
The signal depends on the presence or absence of hydrogen and the degree to which hydrogen is bound within a molecule. For example, in bones, due to the presence of tightly bound hydrogen atoms, the atoms do not align themselves with the magnetic field; thus, they do not produce a usable signal (Katti et al., 2011).
Relaxation
T1 Relaxation
When the magnetization rotates into the xy-plane, the constituent spins obtain energy from the RF pulse. Nevertheless, the spins will gradually return to the original lower energy states, parallel to B0. The process happens when the RF pulse is turned off. The process of returning to the lower levels of energy is called T1, or spin-lattice relaxation. For the spins to return to equilibrium, the energy absorbed is released slowly. When the spins return to equilibrium, the excitation is essentially reset to zero, and the net magnetization rotates back to align with the z-axis, B0, as shown in figure 7. The figure represents the scenario of the net magnetization when RF pulses are removed (Vassiliou et al., 2018; Mastrogiacomo et al., 2019). Following completion of the RF pulse, the net magnetization will have moved to the xy-plane and will start growing back on the z-axis to return to equilibrium. So, the initial growth on the z-axis (panel [a]) will then be greater (panel [b] and [c]). Finally, the magnetization will reach equilibrium (panel [d]) (Vassiliou et al., 2018).
Figure 7: The scenario of the net magnetization when RF pulses are removed.
As not all protons are bound to their molecules in the same manner, some will have stronger or weaker bonds than other protons. This difference leads to a difference in the signal strength between tissues. The difference between the signal strengths constitutes the basis of tissue contrast and forms the substrate for the interpretation of the image. The protons with stronger bonds will allow a more rapid release of energy; that is how those protons will have a shorter T1. This leads to an exponential rise to a maximum, which is the T1 relaxation graph, as shown in graph 1 (Vassiliou et al., 2018). At time t=0, in the same second after the radiofrequency (RF) pulse, there is no magnetization in the z-direction. But immediately after that, the z-axis magnetization starts to recover. The parameter T1 is defined as the time required for the longitudinal magnetization to reach 63% of the initial magnetization. T1 recovery time depends on the type of tissue. Brian tissues are the ones whose 63% of protons return to the longitudinal axis; tissues with a shorter T1 time will be brighter on T1 images and vice versa (Vassiliou et al., 2018).
Graph 1: The T1 relaxation curve.
T2 Relaxation
T2 is the second type of relaxation. In T1, changes happen from the z-axis to the xy-plane. On the other hand, T2 describes changes on the xy-plane. When protons tip back to the z-axis, some are still precessing on the xy-plane. The process of T2 relaxation is shown in figure 8. In the beginning, protons precess in unison; this is called in-phase (panel [a]). All the vectors have different speeds and soon start pointing in different directions (panels (a)–(d)). Although all protons are rotating around the z-axis on the xy-plane, with time, all vectors now point in different directions (panel (e)); this is called out of phase. The process of transforming from utterly in-phase, where all vectors align together, to utterly out-of-phase, where no vector is aligned together, is the T2 relaxation, whilst the total signal from all these vectors pointing in different directions is nearly zero (Vassiliou et al., 2018; Mastrogiacomo et al., 2019).
Figure 8: The process of T2 relaxation.
After the RF pulse is applied, the magnetization transits into the xy-plane; the spins start to precess in phase. Immediately after transmission of the RF pulse ends, the net magnetization vector, or transverse magnetization, rotates in the xy-plane around the z-axis, as shown in figure 6(a). Nevertheless, the direct interaction between the spin magnetic dipole moments, which is known as 'spin-spin interaction’, causes a submicroscopic dispersion in the apparent Larmor frequency of each spin isochromatic, and the magnetization vectors soon start to rotate at different rates (figure 6(b)–(d)). Like T1 relaxation, T2 relaxation causes a decaying effect. However, this happens in the available transverse magnetization detected by the receiver coil. In graph 2, the T2 relaxation graph is shown (Vassiliou et al., 2018).
Graph 2: The T2 relaxation curve.
After the 90 RF pulse, all the magnetization flips into the xy-plane where it is labeled Mxy. Initially, all the spins are in the same phase; nevertheless, immediately after the RF pulse is removed, they start to become out of phase. T2 is the time taken for the total signal to decay to 37% of the original value. T2 usually occurs much faster than T1. Once both the T1 and T2 relaxation processes are finished, the original equilibrium is restored along the main magnetic Vassiliou et al., 2018; Mastrogiacomo et al., 2019).
Image Acquisition
At the time of relaxation that follows the RF pulse, the protons release their excess energy, partly as RF waves. The waves are captured so images can be obtained. This is achieved using a receiver coil, which is sometimes the same as the transmission coil. For safety purposes, the coil is typically placed a distance from the patient. This coil carries peak RF power in the region of 10,000 W in short pulses, which might be dangerous. The coil must be positioned at or near right angles to B0, as the precessing magnetization has an oscillating Mxy component only, which generates a weak oscillating magnetic field in the xy-plane (Vassiliou et al., 2018).
Image creation and display
The RF pulse causes excitation, followed by precession and relaxation, so the RF waves are picked up by the receiver and fed into a computer to produce an image. As shown in figure 9, a flowchart describes the process from sending the RF pulse to viewing the image (Vassiliou et al., 2018).
Figure 9: A flowchart describing the process from sending the radiofrequency (RF) pulse to viewing the image.
Nowadays, computing is based on a two-directional frequency analysis; this is known as the 2D Fourier transform. A computing method, an algorithm, for high speed known as fast Fourier transform (FFT) was essential in the early days of MRI. Nevertheless, this is no longer important given the strong computing power that allows even a simple program for the Fourier transform to be fast enough for images to appear almost instantly after data collection ends (Vassiliou et al., 2018).
T1 and T2 images are different as the contrast differs in different areas. In addition to T1 and T2 images, there are other image types like fluid attenuation inversion recovery (FLAIR), gradient echo (GRE), and diffusion-weighted imaging (DWI). Each one can show a unique detail that is important in a certain case (Copenhaver et al., 2009; Sati et al., 2012; Bammer, 2003).
Advantages of MRI
MRI is one of the best imaging tools nowadays. Some of the advantages are as following (Katti et al., 2011):
• MRI does not use ionizing radiation.
• MRI is non-invasive.
• The contrast resolution of MRI gives it the ability to distinguish adjacent soft tissue from one another.
• MRI makes obtaining a direct, sagittal, coronal, and oblique image possible, which is impossible with radiography and CT.
• Image manipulation is possible.
Disadvantages of MRI
Although MRI is a unique imaging technique, it has many disadvantages, some of which may be unsafe in some cases. Some of the disadvantages are as following (Katti et al., 2011):
MRI equipment is expensive to purchase, maintain, and operate.
• Metallic implants may not be able to have MRIs safely. Thus, metallic personal belongings must be removed. (Peschke et al., 2021).
• MRI images get distorted by metals. Thus, in patients with surgical or stents, the images would be distorted.
• MRI scanners are noisy.
• MRI scans take a long time to scan.
• Claustrophobia, i.e., irrational fear of confined spaces. This happens as the patient is within a large magnet for up to one hour. People generally don't like being in an MRI. Respiration or swallowing may be increased in apprehensive patients (Tischler et al., 2009). In a study in which semi-structured interviews were used to explore the experiences of participants, before, during, and after scanning, healthy volunteers and patients said how they felt. One of the volunteers described his experience by saying: "It was a real shock; I was like oh my god. I thought I was going to explode or something. I was actually quite scared. I was more scared than I thought I would have been. When I first went in I was like… oh my god this is really close. I kind of thought to myself at that point… I started to feel a bit claustrophobic which is really silly cos I’m not claustrophobic really… I didn’t like it." (Healthy volunteer) (Tischler et al., 2009). Some interventions are used to reduce scan-related anxiety levels. Some examples of them are the practice of relaxation exercises, sedation, and the provision of information (Tischler et al., 2009).
Conclusion
This paper has outlined the principles of MRI with an overview of its history. Also, it describes some of the advantages and disadvantages that MRI poses. The MRI is a revolutionary technique in the medical field; its ability to visualize without hurting the patient could contribute to diagnosis development. MRI growth in terms of usage has been growing exponentially. Solving the disadvantages, e.g. the high cost, long time, and fear, would accelerate the use of MRI in the future. It is important to know that it is not utterly safe in every case. Thus, it is recommended that people learn about MRI risks and disadvantages to avoid them.
Acknowledgments
I would like to give a heartfelt thank you to Mr. Arthur Liang, my research mentor at the Columbia VP&S Medicine and Research Summer Program. Mr. Arthur is an incoming freshman at the Massachusetts Institute of Technology (MIT). I would like to thank him for his comments and proofreading.
References
Bammer, R. (2003). Basic principles of diffusion-weighted imaging. European Journal of Radiology, 45(3), 169–184. https://doi.org/10.1016/s0720-048x(02)00303-0
Berger, A. (2002). How does it work?: Magnetic resonance imaging. BMJ, 324(7328), 35. https://doi.org/10.1136/bmj.324.7328.35
Cao, Y., Xu, L., Kuang, Y., Xiong, D., & Pei, R. (2017). Gadolinium-based nanoscale MRI contrast agents for tumor imaging. Journal of Materials Chemistry B, 5(19), 3431–3461. https://doi.org/10.1039/c7tb00382j
Copenhaver, B. R., Shin, J., Warach, S., Butman, J. A., Saver, J. L., & Kidwell, C. S. (2009). Gradient echo MRI: Implementation of a training tutorial for intracranial hemorrhage diagnosis. Neurology, 72(18), 1576–1581. https://doi.org/10.1212/wnl.0b013e3181a411df
Currie, S., Hoggard, N., Craven, I. J., Hadjivassiliou, M., & Wilkinson, I. D. (2012). Understanding MRI: basic MR physics for physicians. Postgraduate Medical Journal, 89(1050), 209–223. https://doi.org/10.1136/postgradmedj-2012-131342
Delantoni, A., & Orhan, K. (2022). Atlas of Dentomaxillofacial Anatomical Imaging (1st ed. 2022 ed.). Springer.
Jahng, G., Park, S., Ryu, C., & Cho, Z. (2020). Magnetic Resonance Imaging: Historical Overview, Technical Developments, and Clinical Applications. Progress in Medical Physics 2020; 31(3): 35-53. https://doi.org/10.14316/pmp.2020.31.3.35
Edwards, J.C. (2009), Principles of NMR. Researchgate. https://www.researchgate.net/publication/268026544_Principles_of_NMR
Geva, T. (2006). Magnetic Resonance Imaging: Historical Perspective. Journal of Cardiovascular Magnetic Resonance, 8(4), 573–580. https://doi.org/10.1080/10976640600755302
Grover, V. P., Tognarelli, J. M., Crossey, M. M., Cox, I. J., Taylor-Robinson, S. D., & McPhail, M. J. (2015). Magnetic Resonance Imaging: Principles and Techniques: Lessons for Clinicians. Journal of Clinical and Experimental Hepatology, 5(3), 246–255. https://doi.org/10.1016/j.jceh.2015.08.001
Jensen, C. D., Duraccio, K. M., Carbine, K. M., & Kirwan, C. B. (2015). Unique Contributions of Magnetic Resonance Imaging to Pediatric Psychology Research: Table I. Journal of Pediatric Psychology, 41(2), 204–209. https://doi.org/10.1093/jpepsy/jsv065
Katti, Girish & Ara, Syeda & Shireen, Dr. (2011). Magnetic resonance imaging (MRI) - A review. Intl J Dental Clin. 3.
Kumar, S., Liney, G., Rai, R., Holloway, L., Moses, D., & Vinod, S. K. (2016). Magnetic resonance imaging in lung: a review of its potential for radiotherapy. The British Journal of Radiology, 89(1060), 20150431. https://doi.org/10.1259/bjr.20150431
Manikkavasakar, S. et al. (2014). Magnetic resonance imaging of pancreatitis: An update. World Journal of Gastroenterology, 20(40), 14760. https://doi.org/10.3748/wjg.v20.i40.14760
Marshall, J., Martin, T., Downie, J., & Malisza, K. (2007). A Comprehensive Analysis of MRI Research Risks: In Support of Full Disclosure. Canadian Journal of Neurological Sciences, 34(1), 11–17. https://doi.org/10.1017/s0317167100005734
Mastrogiacomo, S., Dou, W., Jansen, J. A., & Walboomers, X. F. (2019). Magnetic Resonance Imaging of Hard Tissues and Hard Tissue Engineered Bio-substitutes. Molecular Imaging and Biology, 21(6), 1003–1019. https://doi.org/10.1007/s11307-019-01345-2
Möllenhoff, K., Oros-Peusquens, A. M., & Shah, N. J. (2012). Introduction to the Basics of Magnetic Resonance Imaging. Molecular Imaging in the Clinical Neurosciences, 75–98. https://doi.org/10.1007/7657_2012_56
Moser, E., Stadlbauer, A., Windischberger, C., Quick, H. H., & Ladd, M. E. (2008). Magnetic resonance imaging methodology. European Journal of Nuclear Medicine and Molecular Imaging, 36(S1), 30–41. https://doi.org/10.1007/s00259-008-0938-3
Nelson, C. A. (2008). Incidental Findings in Magnetic Resonance Imaging (MRI) Brain Research. Journal of Law, Medicine & Ethics, 36(2), 315–319.https://doi.org/10.1111/j.1748-720x.2008.00275.x
Palas, J., Matos, A. P., & Ramalho, M. (2013). The Spleen Revisited: An Overview on Magnetic Resonance Imaging. Radiology Research and Practice, 2013, 1–15. https://doi.org/10.1155/2013/219297
Peschke, E., Ulloa, P., Jansen, O., & Hoevener, J. B. (2021). Metallic Implants in MRI – Hazards and Imaging Artifacts. RöFo. https://doi.org/10.1055/a-1460-8566
Sarracanie, M., LaPierre, C. D., Salameh, N., Waddington, D. E. J., Witzel, T., & Rosen, M. S.,(2015). Low-Cost High-Performance MRI. Scientific Reports, 5(1). https://doi.org/10.1038/srep15177
Sati, P., George, I. C., Shea, C. D., Gaitán, M. I., & Reich, D. S. (2012). FLAIR: A Combined MR Contrast Technique for Visualizing White Matter Lesions and Parenchymal Veins. Radiology, 265(3), 926–932. https://doi.org/10.1148/radiol.12120208
Selby, N. M., Blankestijn, P. J., Boor, P., Combe, C., Eckardt, K. U., Eikefjord, E., Garcia-Fernandez, N., Golay, X., Gordon, I., Grenier, N., Hockings, P. D., Jensen, J. D., Joles, J. A., Kalra, P. A., Krämer, B. K., Mark, P. B., Mendichovszky, I. A., Nikolic, O., Odudu, A., . . . Sourbron, S. (2018). Magnetic resonance imaging biomarkers for chronic kidney disease: a position paper from the European Cooperation in Science and Technology Action PARENCHIMA. Nephrology Dialysis Transplantation, 33(suppl_2), ii4–ii14. https://doi.org/10.1093/ndt/gfy152
Serai, S. D., Ho, M. L., Artunduaga, M., Chan, S. S., & Chavhan, G. B. (2021). Components of a magnetic resonance imaging system and their relationship to safety and image quality. Pediatric Radiology, 51(5), 716–723. https://doi.org/10.1007/s00247-020-04894-9
Su, Y., Deng, L., Yang, L., Yuan, X., Xia, W., & Liu, P. (2020). Magnetic resonance imaging for the non-invasive diagnosis in patients with ovarian cancer. Medicine, 99(50), e23551. https://doi.org/10.1097/md.0000000000023551
Tischler, V., Bronjewski, E., O’Connor, K., & Calton, T. (2009). Magnetic resonance imaging (MRI) scanning for research: the experiences of healthy volunteers and patients with remitted depressive illness. Mental Health Review Journal, 14(3), 23–30. https://doi.org/10.1108/13619322200900018
Van Beek, E. J., Kuhl, C., Anzai, Y., Desmond, P., Ehman, R. L., Gong, Q., Gold, G., Gulani, V., Hall‐Craggs, M., Leiner, T., Lim, C. T., Pipe, J. G., Reeder, S., Reinhold, C., Smits, M., Sodickson, D. K., Tempany, C., Vargas, H. A., & Wang, M. (2018). Value of MRI in medicine: More than just another test? Journal of Magnetic Resonance Imaging, 49(7), e14–e25. https://doi.org/10.1002/jmri.26211
Vassiliou, V. S., Cameron, D., Prasad, S. K., & Gatehouse, P. D. (2018b). Magnetic resonance imaging: Physics basics for the cardiologist. JRSM Cardiovascular Disease, 7, 360 204800401877223. https://doi.org/10.1177/2048004018772237
Westbrook, C., & Talbot, J. (2018). MRI in Practice [E-book]. Wiley.