Auscultation | Echocardiography | ECG | MRI | Pedigree

Magnetic Resonance Imaging (MRI)

Medical magnetic resonance imaging (MRI) is a technique for obtaining high-resolution images of various organs within the human body by mapping the distribution of hydrogen nuclei. Compared to echocardiography, MRI gives an undistorted view of the heart; however, the technique is more cumbersome and expensive, and cannot resolve motion over time with as much accuracy as echocardiography. MRI is a non-invasive technique based on the same principles as nuclear magnetic resonance (NMR). The theory behind NMR and the specific details of the technique get rather complicated; what follows is a simplified summary.

Principles of NMR All atomic nuclei with an odd number of protons are positively charged. As nuclei spin, they create micro-magnetic fields around themselves, so that each nucleus resembles a tiny bar magnet with the north and south poles along the axis of spin. Ordinarily, nuclei are oriented randomly so that there is no net magnetic field. However, if a strong and uniform external magnetic field is applied, the nuclei will line up to create a detectable magnetic field.

Spin frequency is defined as:

The gyromagnetic constant is a specific number for each different nuclear species. This means that hydrogen nuclei under a specified magnetic field will spin at a predictable frequency. If the magnetic field changes, the spin frequency changes.

When atomic nuclei are exposed to a radio frequency (RF) electromagnetic wave equal to the Larmor frequency, they absorb energy from the RF wave and become excited. This phenomenon is referred to as resonance. When the RF wave stops, the atoms relax back to the equilibrium state and release absorbed energy to the environment as RF wave emissions. The emitted RF waves can then be detected by sensors.

Application of NMR to medical imaging
So how does one exploit the properties of NMR to obtain two-dimensional pictures? As shown in the figure below, a patient is placed in an external magnetic field oriented in the z-axis (from head to feet), which has variable strength; in other words, the field strength has a smooth gradient so that, for example, it is weakest at the top of the head and strongest at the bottom of the feet. Under these conditions, the proton atoms located near the head have a different Larmor frequency (F1) than those near the feet (F2). The next step is to apply RF waves that match the Larmor frequency of atoms along the length of the body. In our example, the head region would receive RF waves of lower frequency than those to the feet. After the RF waves are stopped, sensors detect the frequency of waves that are emitted back. By selecting a specific frequency, one can view a specific section of the body (or a slice) along the z-axis.

Another magnetic field gradient, perpendicular to the first, is then applied. For example, the second gradient may be weakest near the front of the body and strongest at the back. Once again RF waves are applied and the results recorded.

In this way, a two-dimensional picture can be obtained for any given point in the body. The actual function of detecting the emitted signal and reconstructing it into a visible picture is performed by a computer.

Here is an example of an MRI image of the heart. This particular view is a cross-section of the body near the midline; toward the left is the front of the chest, at the right is the back, and near the top of the picture is the neck.

The heart chambers and the major blood vessels appear as black spaces. AO = aorta, LV = left ventricle, and LA = left atrium. This next picture is at a slightly different plane of section and you can see that the aorta arches over the top of the heart and travels downwards.