Basically, we are interested in knowing one or more of the following.

Simple Detection. Is radiation present? This question usually does not arise in nuclear medicine because one always administers a known radionuclide to the patient. In special circumstances, however, such as contamination of surroundings or personnel, one may be faced with this question. Of course, any presence of radiation will lead to the next question.

Quantity of Radiation. How much radiation or radioactivity is present? In any use of radiation in nuclear medicine, one must answer this basic question, if not in absolute terms, then in relative terms (i.e., with respect to a standard). In dynamic studies, this question is modified as how much radiation or radioactivity is present at a given time? This requires measurement of radioactivity or counting rate as a function of time.

Energy of the Radiation. In nuclear medicine, unlike nuclear physics, one is not interested in knowing the energy of the radiation per se. From practical considerations, however, it is helpful in two ways: to discriminate against undesired events such as scattered γ-rays that may be identified by energy measurement and to identify two radionuclides that may be used simultaneously.

Nature of Radiation. In general, in nuclear medicine, one knows the kind of radiation with which one is involved. However, again in cases of contamination, one may be faced with the task of identifying the type of the radiation.

Not every detector can answer all of these questions. One type of detector can measure each radiation interacting with it individually but not the energy of the radiation [e.g., Geiger-Mueller (GM) counters]. Scintillator detectors can measure both the energy and the individual interaction of the radiation. Ionization chambers cannot be used as counters or to measure energy of radiation but are still useful under many circumstances, such as measurement of radiation “exposure.” Exposure depends on both the energies and the number of radiations interacting in the detector and is a useful concept in radiation protection from external x- or γ-ray emitting radiation sources (see Chapter 18).

The answer to this complex question primarily depends on the use one wishes to make of the radiation detector. However, the following properties of a detector are of significance in radionuclidic imaging, which is the primary activity in nuclear medicine.

Intrinsic Efficiency or Sensitivity. Intrinsic efficiency E_i of a detector is the measure of its ability to detect radiation and is generally defined as the ratio of the number of rays of a given radiation (a, β, or γ) detected to the number of rays incident on the sensitive volume of the detector:

A value of 0.5 (50%) for the intrinsic efficiency means that only one-half of the rays incident on the sensitive volume of the detector is detected and the other half passed without any interaction within the sensitive volume. The higher the intrinsic efficiency of a detector, the better it is for use in nuclear medicine. Higher intrinsic efficiency allows a shortened imaging time, a reduced radiation dose to patients, or both. Intrinsic efficiency of a detector primarily depends on the linear-attenuation coefficient, µ(linear), and the thickness of the sensitive volume. As discussed in Chapter 6, p. 63, µ(linear) itself depends on the γ ray energy as well as the nature of the material (atomic number and density).

Dead Time or Resolving Time. Dead time or resolving time (used interchangeably here) is a measure of the ability of a detector to function accurately at high count rates encountered when large amounts of radioactivity are used. For any detector, there is a small but finite interval between the time when a ray interacts with a detector and the time when the detector responds, and the event is recorded. This interval is known as “dead time” or resolving time of the detector. What happens if a second ray arrives and starts interacting with the detector while it is still processing the first ray? The response of a detector under such circumstance is broadly described under two categories: paralyzable and nonparalyzable. Both mechanisms may be operative in the same detector.

In the paralyzable situation when the second ray arrives within the dead time of the detector, the detector becomes insensitive for another time interval equal to the time of arrival of the second ray. For example, if the dead time of a detector is 100 µs and the second ray arrives after 30 µs of the arrival of the first, the detector becomes insensitive for a total of 30 + 100 = 130 µs, provided, of course, no third ray arrives within this interval. In that case, the interval will be extended further depending on the time of arrival of the third ray and so on. Thus, depending on the count rate, which determines the average time interval between the arrivals of two successive rays, in the paralyzable case a large number of rays can be lost without registering in the detector.

In the nonparalyzable case, the insensitive period of the detector is not affected by the arrival of the second ray. The second ray or others arriving in this insensitive period are lost. Figure 8.1 shows the effect of dead time on various count rates for a paralyzable
and nonparalyzable detector. For a nonparalyzable detector, the detector response (observed count rate) at first increases linearly with the true count rate but then reaches a plateau. On the other hand, for a paralyzable detector, the initial part of the response is the same, but after reaching a peak, the observed count rate actually starts decreasing with an increase in the true count rate. In other words, for paralyzable-type counters, the same observed count rate is obtained with two different true count rates (or amounts of radioactivity), one low and one high; this is a serious pitfall to be aware of.

Ideally, the detector should have as short a dead time as possible because it will detect very high count rates without significant losses. For the count rates encountered in nuclear medicine from the routine dosage of the current radiopharmaceuticals, a system dead time of about 10 µs is acceptable. However, for fast dynamic imaging of the heart where higher count rates are desirable, system dead time shorter than 10 µs (2 to 3 µs) is essential.

Energy Discrimination Capability or Energy Resolution. The ability of a detector to distinguish between two radiations of different energies (e.g., two γ-rays of different energies) is known as its energy discrimination capability or energy resolution. Full-width at half-maximum (FWHM) is commonly used as a measure of energy resolution and is defined later in this chapter. It represents the minimum difference necessary between the energies of two γ-rays if they are to be identified as having different energies. If the FWHM of a detector is 20 keV, then two γ-rays with a difference in energy <20 keV cannot be distinguished by this detector. A lower value of FWHM indicates a better energy discrimination capability. Another way to look at FWHM is to consider it as a measure of error in the energy determination of an x- or γ-ray by a detector. The lower this error is, the more accurate the detector will be in measuring γ-ray energy and therefore in detecting smaller changes in the energy of x- or γ-rays.

Other Considerations. Because most detectors use electronic components whose behavior quite often is susceptible to change with fluctuations in line voltage and environment temperature, detectors whose response is not appreciably affected by these fluctuations are preferred. Detectors should also be portable (depending on their desired use), simple to operate, and, of course, inexpensive.

With this background in mind, various types of radiation detectors are discussed: (A) gas-filled detectors, (B) scintillation detectors, and (C) semiconductor detectors. Their main characteristics and uses are summarized in Table 8.1.

As discussed in Chapter 6, the first action of high-energy radiation on matter is the production of ionization. In general, it is not possible to measure the amount of ionization produced in matter except in gases and some solids, known as semiconductors. The measurement of the amount of ionization produced by radiation in a small gas volume is the basis of three types (i.e., ionization chambers, proportional counters, and GM counters) of gas-filled detectors.

Mechanism of Gas-Filled Detectors. To explain the mechanism of detection of gas-filled detectors, let us consider what happens when ionization is produced by a ray or particle in a gas volume enclosed between two electrodes with an applied voltage V (Fig. 8.2). When the applied voltage V between the two electrodes is zero, the ion pairs produced by the radiation recombine to form neutral atoms or molecules, and no current flows in the circuit. However, under the influence of the electric field that exists between the two electrodes when V is greater than zero, some ion pairs reach the electrodes and a transient current (current pulse) is produced in the circuit. The amount of current produced depends on several factors, such as the applied voltage V; distance between the two electrodes; type of gas; volume, pressure, and temperature of the gas; and geometry and shape of the electrodes. The most important parameter, however, is the applied voltage V between the two electrodes. The amount of the current produced by a single radiation in a typical gas-filled detector as a function of voltage V is shown in Figure 8.3. The dependence of the current on voltage is complex. There are five distinct regions that need explanation. Keep in mind that if no ionization is produced by the radiation in the gas volume, the current will be zero irrespective of the voltage applied.

Region I or Recombination. The voltage is low so that some ion pairs produced by a ray or particle (called primary ion pairs) are still able to recombine and form neutral atoms or molecules. In other words, there is an incomplete collection of the primary ion pairs by the electrodes. As the voltage increases, more primary ion pairs are collected by the electrodes and more current flows.

Region II or Ionization Plateau. The voltage is now sufficiently high to attract all primary ion pairs produced by a ray or particle to the electrodes, and there is little recombination. The amount of the current does not change appreciably with an increase in the voltage because all primary ion pairs produced by the radiation have been collected. The amount of current produced by a single ray or particle (total number of primary ion pairs) is still small and cannot be detected unless integrated for interaction of many rays or particles.

Region III or Proportional Region. The voltage is even higher. This is not only enough to attract all primary ion pairs to the electrodes but is also sufficient to provide enough energy to some primary ion pairs that now produce secondary ion pairs through collisions with neutral atoms and molecules of the gas. The amount of secondary ion pairs produced depends on the energy acquired by the primary ion pairs, which depends on the voltage. As a result, the amount of current produced by a radiation increases with increase in voltage. The current is now sufficiently high for detection in response to a single ray or particle interaction in the gas volume. It is also proportional to the energy of the radiation. However, as the voltage increases further (unmarked region between III and IV), this relationship with energy becomes nonlinear.

Region IV or GM Plateau. The voltage is so much higher that the primary ion pairs produced by a ray or particle acquire large energy to produce a large number of secondary ion pairs and excitations, which then strike the metallic electrodes and more neutral gas molecules to produce more ionizations and excitations. The de-excitation of the molecules produces ultraviolet light, which produces more ionization. The result is that the total gas volume is in a temporary state of discharge, and the amount of current produced is more or less independent of the voltage and the energy of the radiation.

Region V. The voltage is so high that radiation is not necessary to produce discharge. Under the influence of such a high electric field, the electrons are pulled out from the atomic shells, the atoms and molecules become ionized, and a discharge may be established even without radiation.

Types of Gas-Filled Detectors. There are three types of gas-filled detectors: (1) Ionization detector (chamber), dose calibrator is a special purpose ionization detector; (2) Proportional detector (counter); and (3) GM detector (counter).

Ionization Detector (Chamber). Ionization detectors are one of the earliest types of gas-filled detectors used for radiation measurement. The operating voltage in these detectors lies in region II of Figure 8.3. In this region, small fluctuations in the applied voltage do not affect the current significantly. As a result, ionization chambers are highly stable and reliable. These can be made in a variety of shapes and sizes and are relatively inexpensive. Their main disadvantages are a poor intrinsic efficiency for x- or γ-rays and a lack of energy discrimination. Because the amount of current produced in response to a single ray or particle is small, these cannot be used as counters. Therefore, their principal use is in the measurement of radiation exposure such as that encountered in diagnostic and therapeutic radiology. They are principally used in nuclear medicine as “dose calibrators” to measure radioactivity in the microcurie-to-curie range. Occasionally, these may be used to measure exposure to individual radiation workers in the form of “pocket dosimeters,” as discussed in Chapter 18.

Dose Calibrator. Dose calibrators are, generally, cylindrically shaped ionization chambers filled with a rare gas such as argon at high pressure (˜20 atmospheres). High pressure increases the density of the gas, which increases the intrinsic efficiency of the ionization chamber. There is a small hole along the axis of the cylinder, so that the radioactive sample whose radioactivity is to be measured can be inserted close to the center of the ionization chamber (Figs. 8.4 and 8.5). This type of geometric arrangement increases the overall sensitivity of a detector and is known as 4π arrangement (see Chapter 9). The outside walls of the chamber are appropriately shielded to minimize the interference from radioactive sources outside the chamber.

The principle and operation of a dose calibrator are simple. The current produced in the ionization chamber by a given radioactive source in a given geometric arrangement is directly proportional to the amount of the radioactivity of the source. However, different radionuclides with the same amount of radioactivity produce different amounts of current. This difference in current produced by the same radioactivity of different radionuclides arises mainly from the differences in emission frequency n_i and the energy E_i of γ-rays emitted by them, as well as the β branching ratios, all discussed in Chapter 2. Therefore, before an ionization chamber can be used as a dose
calibrator, it needs to be calibrated for the desired radionuclides individually. The initial calibration is performed by measuring the amount of current produced for 1 mCi of a given radionuclide. This process is repeated for all desired radionuclides. Sources of known radioactivity (with 5% or better accuracy) can be obtained from the National Institute of Standards and Technology (NIST), which determines the radioactivity of these sources accurately by sophisticated and time-consuming methods. However, for radionuclides, for which sources of known radioactivity are not available, as may be the case for short-lived radionuclides in positron emission tomography (PET), a complex procedure is followed that is out of the scope of this book. Once the calibration factors are known, the unknown radioactivity of a given radionuclide is easily obtained by dividing the current produced by the unknown radioactivity with the calibration factor for that radionuclide.

In commercial dose calibrators (Figs. 8.4 and 8.5), the ionization chamber is connected to a digital current meter, and the calibration factors for a number of radionuclides are predetermined and stored so that the results of the unknown radioactivity are directly displayed in microcurie, millicurie, or curie (or MBq, etc.) when the appropriate button for the given radionuclide is selected. With these calibrators, however, it should be remembered that when we press a button labeled “^99mTc,” all we are doing is recalling the calibration factor of ^99mTc. We are not discriminating against other radionuclides. A ^99mTc source in a dose calibrator will display radioactivity at other radionuclide settings, but these will not be the correct radioactivity of the ^99mTc sample. As a result, dose calibrators cannot be used to determine the radioactivity of a mixed (containing more than one radionuclide) sample except as described under “⁹⁹Mo Breakthrough Measurement” shortly.

Quality Assurance of a Dose Calibrator. To ensure the proper operation of a dose calibrator, its linearity (i.e., the current is proportional to the radioactivity for a given radionuclide) and accuracy should be measured yearly. Accuracy should be better than 10%. For routine assurance, a constancy check should be performed daily by measuring the radioactivity of a standard source consisting of a long-lived radionuclide such as ¹³⁷Cs or ⁵⁷Co. The measured radioactivity should not vary more than 10% from the radioactivity of the standard source, corrected for the decay. It must also be noted that the calibration factors are valid only for a given geometric setup, source volume, and source container. If the shape or type of the source container or the volume of the source is changed appreciably, the calibration factor will change and therefore should be measured again.

⁹⁹Mo Breakthrough Measurement (An Unusual Application of a Dose Calibrator). Although a dose calibrator cannot measure accurately the radioactivities of two radionuclides when present simultaneously in a sample, it can be used to measure the radioactivity of ⁹⁹Mo breakthrough in the ^99mTc eluate obtained from a generator by using an ingenious method. It involves the use of a lead container of a wall thickness sufficient to attenuate essentially all γ-rays (140 keV) emanating from a typical ^99mTc radioactivity obtained from a generator. This thickness of lead is not enough to attenuate high-energy ⁹⁹Mo γ-rays (740 and 778 keV) to any great extent (Question 8, Chapter 6 illustrates this point). Lead containers of appropriate thickness are usually provided by the manufacturers of dose calibrators. To measure ⁹⁹Mo breakthrough, two measurements are made on the vial containing ^99mTc eluate: on the ^99mTc setting without the lead container, which gives the ^99mTc radioactivity; and at the ⁹⁹Mo setting with the vial in the lead container, which, when multiplied by the attenuation factor for ⁹⁹Mo γ rays (provided by the manufacturer), yields the ⁹⁹Mo radioactivity. Because the ^99mTc radioactivity in the eluate is typically ˜100,000 times the ⁹⁹Mo radioactivity (parent breakthrough), error in the ^99mTc radioactivity in the presence of a small amount of ⁹⁹Mo radioactivity is small and therefore neglected. Division of the ⁹⁹Mo radioactivity (in microcurie) by the ^99mTc radioactivity (in millicurie) yields the ⁹⁹Mo breakthrough in the appropriate unit.