In order to evaluate radiation exposures such as those produced by advanced imaging technology (AIT) systems, and to make it possible to estimate health effects that might result from those exposures, it is necessary to have a way to measure radiation exposure and to account for the different biological effectiveness of different types of ionizing radiation. Quantification of radiation exposure is complicated because a wide range of ionizing radiation is found in nature, and the interactions that those radiation types have with matter are complex. The International Commission on Radiation Units and Measurements (ICRU) has developed a self-consistent set of radiation quantities and units to resolve this problem (ICRU Report 85a, 2011). Because the AIT systems under consideration here use only photons with relatively low energy (less than 120 keV), this chapter will summarize only the quantities and units that are relevant to the determination and reporting of exposures to low-energy X-rays. The quantities that are needed for the work in Chapters 6 and 7 include the following:
- A measure of the energy spectrum of the photons,
- The amount of energy lost in a material by photons,
- The amount of energy deposited in tissue, and
- A quantity suitable for comparison with radiation protection limits when the type of radiation is different or when only a portion of the body is exposed to radiation.
The energy spectrum of photons incident on an exposed individual is related to the probability of each type of photon interaction and therefore affects the overall attenuation of the photon beam with depth. Consequently, it is essential to characterize the photon spectrum in order to determine the energy deposited in different parts of the body. The X-ray emission from conventional X-ray tubes is of much lower brightness than many other sources such as synchrotrons. Although it is possible to determine the spectra produced by X-ray tubes using semiconductor or other energy-sensitive detectors, when methods for characterization of X-ray beams were first developed, techniques for measuring X-ray intensity were used to characterize the photon spectrum, not spectroscopic principles.
In practice, determination of the intensity of photons emitted by an X-ray tube is generally accomplished by determining the accumulation of electrical charge collected in ionization chambers placed in the beam. This is possible because the photon energy required to produce an ion is nearly independent of the photon energy. Measurements of beam attenuation are used to confirm assumptions about the photon energy spectrum. In medical practice, knowledge of the tube’s high voltage, anode material, the angle of the beam relative to the anode surface, and the amount of inherent and external filtration are sufficient to characterize a spectrum for most imaging applications.
The term generally used to characterize the spectrum of photons emitted by an X-ray tube is beam quality. It refers to the penetration capability of an X-ray beam emerging from a source. In clinical-diagnostic practice, beam quality is expressed as the first and second half-value layer (HVL) thicknesses, HVL1 and HVL2, respectively. The first half-value layer reduces the intensity of the incident beam by 50 percent, and adding the second half-value layer reduces the incident beam by another factor of two for a combined attenuation of 75 percent. If HVL1 and HVL2 are known, together with tube voltage, it is possible to estimate the energy spectrum of the emerging beam using established spectral models.
In order to quantify X-ray beam intensity, a precise measure of the intensity is needed. The standard measure of energy transferred to matter by indirectly ionizing radiation (X rays and neutrons) is air kerma.1 In the case of X rays, kerma is the kinetic energy transferred from photons to charged particles in a volume of
1 When a quantity to characterize energy lost by radiation, as opposed to energy deposited in matter, was first introduced, it was identified by the acronym KERMA, kinetic energy released in matter. More recently, kerma has been adopted as the name of the quantity.
material, divided by the mass of material in that volume. It is typically expressed in gray (Gy) where 1 Gy = 1 J/kg.
Although other types of detectors can be used, gas-filled ion chambers are generally preferred for measuring kerma in the range of beam intensities produced by X-ray tubes. The radiation traversing the chamber produces positive and negative ions and free electrons, which are collected at two electrodes (typically concentric cylinders or parallel plates) by applying a potential difference between them. Ideally, all ions and electrons are collected. However, this is usually not achieved, and some corrections are generally necessary. Even the best chambers have small regions where the electric-field strength is reduced. This results in a local increase in recombination of electrons and ions in the gas and reduces the collected charge. Recombination can be reduced by reducing the density of the gas, but this negatively impacts sensitivity. Recombination can also be reduced by increasing the voltage difference between the plates. If the collected charge is found to be essentially independent of applied voltage for a range of voltages, it is a good indication that all of the charge that can be collected has been collected.
Ion-chamber design is also relevant. Most ion chambers necessarily have solid walls to define the gas volume where ionization is measured. The walls also provide secondary electrons that ionize the gas. In addition, a solid wall will preferentially attenuate low-energy photons. As a result, measurement of kerma is strongly dependent on the photon spectrum, and ion chambers must be calibrated in a radiation field that has the same quality as the radiation fields they will be used to measure. Size also matters, because sufficient charge must be produced to be measurable by the associated electronics. For example, a 1 cm3 chamber exposed to 1 nGy would produce approximately 200 ion pairs, or 3.2 × 10–17 coulombs. This is much less than can be accurately measured even by high-quality electrometers. However, an 1,800 cm3 ion chamber exposed to 1 nGy would produce 5.7 × 10–14 coulombs, which can be measured.
To minimize measurement uncertainty and provide a traceable standard, the National Institute of Standards and Technology (NIST) maintains precision free-air ion chambers to calibrate ion chambers in terms of air kerma for a variety of X-ray spectra.2 The free-air ion chamber is a specialized instrument with no walls defining the sensitive volume of air. It measures the electric charge produced in a volume of air defined by two electrodes and a carefully collimated beam of photons. Because the radiation must be limited to a precisely defined beam, the free-air ion chamber cannot be used to measure kerma in a typical radiation environment. However, it can be used to cross-calibrate other ion chambers in well-defined beams. By sequentially placing the free-air reference chamber and the test cham-
bers in the same beam, NIST can determine the calibration factor for a given test chamber. Because each chamber has its own energy dependence, the calibration factor is specific to the spectrum used for the calibration. However, NIST provides a wide range of different spectra for calibration, as determined by the operating conditions of the X-ray tube and additional filtration.
To predict the biological consequences of irradiation, kerma is not enough. The actual amount of energy deposited in tissue is more relevant because it is assumed that this energy initiates the biochemical processes that lead to the observable effects of irradiation. Because some of the electrons liberated by photon irradiation have ranges that are larger than the dimensions of cells in tissues, the energy deposited in tissue may not exactly equal the energy lost by photons. As a result, the absorbed dose, D, is defined as dε/dm, where dε is the element of energy absorbed from ionizing radiation by an element of mass dm.3D and kerma have the same units but differ in that kerma describes the kinetic energy transferred to electrons without regard to their final destination, and D describes energy deposited without regard to origin. Thus, D does not include the energy escaping the volume but does include energy depositions within the volume from electrons initiated outside of the volume. Kerma and absorbed dose are numerically equivalent under the conditions of charged particle equilibrium (CPE). While proportional to the intensity of the incident X-ray beam, neither gives the intensity of the beam directly.
For purposes of radiation protection, exposure to different types of radiation must be described in a way that can be compared to a single exposure limit, independent of the type of radiation or its spatial distribution in the body of the exposed individual. The quantity “effective dose,” E, has been established by the ICRP for this purpose. E is the sum of doses to specific tissues or organs multiplied by the dimensionless weighting factors WR (the radiation weighting factor) and WT (the tissue weighting factor), which describe the different sensitivities of the different organs to different types of radiation (see Box 4.1). As with kerma and absorbed dose, the units of effective dose are J/kg, but because E is not a measurable physical quantity, its unit is given the special name sievert (Sv) instead of using the physical measurable quantity gray (Gy).
Several approaches can be used to determine the effective dose. Because expo-
3 International Commission on Radiation Units and Measurements, ICRU Report 46: Photon, Electron, Proton and Neutron Interaction Data for Body Tissues, Journal of the ICRU, 1992.
Dosimetry Quantities and Units
The International Commission on Radiation Units and Measurements (ICRU) has developed a radiation measurement system using both stochastic and non-stochastic quantities.1 The non-stochastic quantities are defined as expectation values at a point in space, that is, in the differential form. Since the non-stochastic quantities are used for radiation protection their definitions are given here.
Kerma, K, is defined as1:
where dEtr is the sum of initial kinetic energies of all charged ionizing particles liberated by uncharged ionizing particles in a volume of material having a mass dm.1 Kerma has dimensions of energy divided by mass. It is described by the quantity gray (Gy) where
Absorbed dose, D, is defined as1
where d ε is the mean energy imparted by ionizing radiation in a volume of material of mass dm.1 Absorbed dose has dimensions of energy divided by mass. It is also described by the quantity Gy. Absorbed dose and kerma differ in the sense that kerma represents the kinetic energy transferred to electrons, whereas absorbed dose represents the energy absorbed. One distinction is that kerma is based on the kinetic energy of electrons originating in a volume without regard to the final destination. Absorbed dose is based on energy deposition in a volume without restrictions on the origins or final destinations of the electrons. Kerma and absorbed dose are numerically equivalent under the conditions of charged particle equilibrium.
Effective dose,2E, is the tissue weighted sum of equivalent doses in specific tissues and organs of the body, and is given by the expression
sure to different organs depends on the attenuation of the beam at their locations, the most direct approach is to use a material phantom. This simulates a person of the desired height and weight and contains a large number of small dosimeters at appropriate locations. The dosimeters are used to evaluate the average absorbed dose to each of the organs. For X rays, the radiation weighting factor (WR) is 1.0, so the effective dose is obtained by applying the appropriate tissue weighting factors
where DT,R is the absorbed dose from radiation of type R averaged over a tissue or organ T, and wR is a weighting factor for radiation of type R. For photons of all energies, wR = 1.
The tissue weighting factor, wT, is the factor by which the absorbed dose in a tissue or organ T is modified to represent the relative contribution of that tissue or organ to the total health detriment from a uniform irradiation of the whole body. It is normalized to
Tissue weighting factors are based on epidemiological studies of cancer induction, genetic alterations following exposure to radiation, and judgment. Furthermore, they represent mean values for humans averaged over both sexes and a span of ages. Effective dose is intended for use as a protection quantity derived from reference values and therefore is not recommended for epidemiological evaluations, nor should it be used for detailed assessment of exposure and risk to specific individuals.
Reference effective dose,3EREF, is determined for full-body AIT systems from measurements of the half-value layer (HVL) and air kerma (Gy) as outlined below. It is given specifically by
where EREF is the reference effective dose in sieverts (Sv), Ka is the measured air kerma per screen in grays (Gy), and C (in Sv/Gy) is given by C = 0.125 × HVL in millimeters of aluminum or 1.14, whichever is smaller.
1 International Commission on Radiation Units and Measurements, ICRU Report 85a: Fundamental Quantities and Units for Ionizing Radiation, Journal of the ICRU 11(1), 2011.
2 International Commission on Radiological Protection, ICRP Publication 103: The 2007 Recommendations of the International Commission on Radiological Protection, Annals of the ICRP 37(2-4), 2007.
3 The American National Standards Institute/Health Physics Society standard ANSI/HPS N43.17-2009, “Radiation Safety for Personnel Security Screening Systems Using X-Ray or Gamma Radiation,” is available at the Health Physics Society website at http://hps.org/hpssc/index.html.
to these data and adding the results. The process is difficult and time-consuming, with accuracy limited by the accuracy of the individual dosimeters and the number of dosimeters used to determine the average dose in specific tissues.
The alternative to direct measurement is computation. Here, radiation-transport calculations determine the absorbed dose at specified locations in a computational phantom. The computational phantoms originally used for this
purpose were collections of geometric shapes (cylinders, spheres, and intersecting planes) arranged to simulate a human body. Recently, phantoms of considerably improved detail and accuracy have been developed using computed-tomography images of individuals.
The American National Standards Institute (ANSI) has adopted the concept of a reference effective dose, EREF, for specifying exposures produced by X-ray AIT systems (see Box 4.1). EREF is equal to the product of the air kerma of the beam and a function of its HVL. The justification for EREF is based on the observation that “the effective dose for monoenergetic photons and for all the spectra having at least 1 mm of aluminum total filtration follows a relatively close-grouped pattern as a function of HVL.”4 The ANSI/Health Physics Society (ANSI/HPS) standard N43.17-2009 specifies that the reference effective dose from general-use AIT systems (such as AIT equipment) “shall not exceed 250 nSv (25,000 nrem) per screening” where a screening consists of an anterior and a posterior scan.5 This recommendation is generic and does not specifically focus on specific at-risk populations.
To compare E produced by X-ray backscatter AIT systems to the maximum effective dose specified by National Council on Radiation Protection and Measurements (NCRP) and the ANSI/HPS N43.17-2009 standard, it is necessary to determine the average absorbed dose in each of the organs for which a weighting factor has been defined. To do this, it is necessary to consider the diffusion of electrons and the effect of boundaries. Diffusion is the process by which higher concentrations of electrons expand across a boundary into regions of lower concentrations. If the system is in local CPE, meaning that the local rates of in- and out-diffusion of energy are equal, then D is equal to the kerma, and its evaluation is straightforward. However, CPE does not occur near boundaries between materials of different atomic compositions. Of particular concern is the dose to the skin, because in most cases it is adjacent to air or in contact with another material and hence falls into this category.
For a beam of a given intensity, the concentration of secondary electrons cre-
4 The ANSI/HPS N43.17-2009 standard, “Radiation Safety for Personnel Security Screening Systems Using X-Ray or Gamma Radiation,” is available at the Health Physics Society website at http://hps.org/hpssc/index.html.
5 ANSI/HPS N43.17-2009, p. 6.
ated in any material is proportional to the kerma in that material. Therefore, at a boundary between dissimilar materials, the concentration of secondary electrons is higher on one side than the other. As a result, electrons in the material with the higher kerma will diffuse across the boundary to the material with the lower kerma. Therefore, near the boundary, D will exceed the value calculated from kerma alone in the material with the lower kerma, and vice versa. This difference will die out on either side over distances of the order of the characteristic diffusion lengths for secondary electrons in the materials. The scale of these distances can be approximated as the range of the most energetic secondary electrons in the material, which for a 50 keV electron in tissue is 44 µm. Because most secondary electrons have much lower energies and therefore much shorter ranges, equilibrium between D and kerma is typically established well within 50 µm of the surface. Because the beam intensity decreases approximately exponentially with depth, diffusion causes D to be slightly different than the kerma, but this is usually a minor effect and can be ignored. Because nearly all radiation-sensitive cells are deeper than 50 µm in the body, D is essentially equal to the kerma for all radiation-sensitive cells.
If a material with a high energy transfer coefficient (µtr/ρ,) is in contact with the skin, D at the surface of the tissue can be much higher than is the kerma. For example, at 30 keV, the coefficient µtr/ρ of aluminum is 5.6 times that of muscle. The ratio for silicon is 7.4, and for calcium 23.4. If the skin is covered by a thin layer of one of these materials, energy transfer across the skin–material boundary needs to be taken into account.
Photon scattering also influences the kerma near boundaries between materials of different atomic compositions. Backscattered photons increase the kerma relative to that produced by the incident photons alone. The magnitude of this photon buildup depends on the geometry of the radiation field and of the object irradiated but is generally not large. It is included in the results of the Monte Carlo codes6 used to calculate effective dose in a phantom.
6 Monte Carlo codes are a type of computational algorithms that rely on repeated random sampling to obtain numerical results; typically, one runs simulations many times over in order to obtain the distribution of an unknown probabilistic entity.