PULSE SEQUENCE AND IMAGE CONTRAST
In this section we will discuss the most common basic MR imaging pulse sequence, spin-echo imaging, and the various factors that affect image contrast in this pulse sequence.
A number of parameters influence MR image contrast. They can be conveniently divided into two general categories: inherent tissue parameters and user-selectable imaging parameters.
Inherent tissue parameters, that affect MR signal intensity (blackness or whiteness) of a particular tissue are primarily (a) the relative proton (i.e., hydrogen) density of that tissue and (b) its inherent T! and T2 relaxation times (other factors such as flow also affect signal intensity but a detailed discussion of this is beyond the scope of this primer). The contrast between two adjacent tissues (for example, between a brain tumor and normal white matter) depends on the difference between the proton densities, T1s and T2s of the two tissues. This so-called "conspicuity" (lesion verses background tissue signal) can be maximised by appropriate manipulation of user-selectable parameters. Inappropriate pulse sequences can diminish the signal difference between lesion and background tissue, making lesion detection difficult.
User-selectable parameters, that affect image contrast and are under control of the radiologist include choice of pulse sequences, flip angles, section thickness and field of view, matrix size, number of excitations and use of exogenous contrast agents.
The most commonly used pulse sequence in MR imaging is a spin echo sequence. Here 900 Rf pulse is sent into the patient to tip the tissue magnetisation into the transverse plane. While an image could theoretically be obtained at this instant, in real life some time must elapse between the 900 pulse and signal measurement (phase and frequency encoding for spatial localisation must be performed). Since transverse de-phasing (T2 decay) occurs quickly, a 1800 re-phasing pulse is applied a few milliseconds or tens of milliseconds after the 900 pulse (this eliminates reversible or T2* decay and leaves only real T2 relaxation to affect signal intensity). The output or signal echo is measured an equal time (a few seconds to tens of milliseconds) after the 1800 re-phasing pulse is applied. This sequence of RF input/re-phasing pulse/echo is repeated many times in a single study. The time between the repetition of each sequence is the time interval between 900 pulses. This is called TR or "Time to Repetition". The time between the middle of the 900 pulse and maximum signal output (or echo) from the patient is called the TE or "Time to Echo". The signal intensity of a given volume of tissue or voxel in SE imaging is expressed by the following relationship:
Sse (TE, Tr) = N[H] e-TE/T2 (1-e-TR/T1)
This looks complex but all we need to observe is that TE appears with T2 and therefore controls the degree of T2 weighing, TR appears with T1 and therefore controls the degree of T1 weighing and spin density is always present as a factor in determining signal intensity. Thus, simply by varying TR and TE it is possible to obtain an image that is predominately T1-, T2- or spin-density weighted. In general, T1- weighted images provide the best anatomic detail while T2- weighted studies are better for detecting disease, such as discriminating lesions from background normal tissues.
Contrast between any two tissues is the difference in measured signal output at a given point in time. Note that if TR and TE are both short (as occurs in a T1-weighted scan) white matter will appear brighter than CSF. If TR and TE are both long (as in a T2-weighted scan) CSF will appear brighter than brain. If signal output is measured at precisely the point where the decay curves of the two tissues cross, they will be indisntinguishable from each other because they are isointense. This concept has a very practical implications as it is possible to make lesions "disappear" (i.e., become isointense compared to their background normal tissue) if inappropriate Tr and TE values are chosen. It is extremely important to select Tr and Te for optimum image contrast and lesion "conspicuity".
In summary, by varying TR and TE in spin-echo MR, image contrast can be manipulated to display varying intensities based primarily on T1, T2 or proton density.
NMR IMAGING
Spatial coding The wavelength of protons used in NMR imaging are of such great length (75m at 4MHz) that projection imaging as we are familiar with in other forms of imaging used in medicine is completely impossible. All forms of NMR imaging use some form of spatially coded frequency discrimination to allow formation of a picture. This may sound formidable, but conceptually it is quite simple and understanding the basic idea is crucially important.
As we discussed earlier, the frequency of magnetic resonance for a specific nucleus is dependent on the strength of magnetic field. This situation is analogous a vibrating string. As the tension on a string is increased, the frequency of resonance increases. We could, of course, also change the resonance frequency by changing the linear mass density or size of the string. Increasing the magnetic field strength is equivalent to changing the tension on the nucleus and results in increases in the resonant frequency, while changing the size of the string is equivalent to changing from one nuclear species (magnetogyric constant) to another. In NMR imaging frequencies and intensities are sufficiently different that we normally topical only one type of the nucleus at a time.
NMR imaging apparatus
Zeugmatography. The NMR imaging procedure we have been discussing was named Zeugmatography by Paul Lauterbur in 1973. The green word "zeugma" means to yoke together. NMR imaging yokes together a magnetic field and radio frequency photons to provide a picture of the internal structures of the human body. There are many principles in applying this general zeugmatography principles as shown in Table below, which lists general categories for variations in the excitatior and emission detection phases.
Simpule Excitatior Simple Detection
Sensitive Excitatior Sensitive Detection
Preparative Excitatior Prepared Detection
More than twenty types of imaging have been proposed and tested to date with only a partial listing of these given in Table below.
Table : NMR imaging methods
The multiplicity of potential imaging modulities raises many questions for equipment manufacturers and clinical users. To date no method has proved universally advantageous. The solution of these problems will require testing if clinical application situations and certainly some compromises will be necessary. It is very unlikely that any one machine will provide all the solutions. In spite of the variability there are common elements to all NMR scanners.
General description. A schematic description of the NMR scanner will consists of six major subsystems as shown in the figure below; primary magnet, computer, radio equipment, magnetic gradient, data storage and display subsystems.
The role and function of computer, data storage and display systems are identical in NMR imaging to the role in CT imaging and therefore familiar to most medical imaging personnel. We therefore refer you to existing literature concerning these devices and limit our brief discussion to the remaining three subsystems.
