RADIOLOGICAL PHYSICS Raphex+ Preface @ The RAPHEX 2007 Exam Answers book provides a short explanation of why each answer is correct, along with worked calculations where appropriate. An in-depth review of the exam with the physics instructor is encouraged. In cases where more than one answer might be considered correct, the most appropriate answer is used. Although one exam cannot cover every topic in the syllabus, a review of RAPHEX exams/answers from three consecutive years should cover most topics. We hope that residents will find these exams useful in reviewing their radiological physics course. RAPHEX 2007 Committee Copyright © 2007 by RAMPS, Inc., the New York chapter of the AAPM. All rights reserved. No part of this book may be used or reproduced in any manner whatsoever without written permission from the publisher or the copyright holder. Published in cooperation with RAMPS by: Medical Physics Publishing 4513 Vernon Boulevard Madison, WI 53705-4964 1-800-442-5778 E-mail: [email protected] Web: www.medicalphysics.org LeGeneral “ Answers * Gl. E G2. D By definition, 1 Sv = 100 rem. G3. D Exposure is equal to the total charge of ions of one sign produced in a unit mass of air. Units were formerly roentgen, R, and are now C/kg in the SI system. G4. C It should be 1.0 Sv/Quality Factor. G5. D G6. D G7. E This can be found from Einstein’s equation: E = me’. G8. B A stable isotope of hydrogen with one proton and one neutron in the nucleus is called deuterium. Go A Glo. E Gil. D GI2. C Gli3. E Electron mass is equivalent to 0.511 MeV, using Einstein’s equation: E = me’, Gl4. A GIs. C Isotopes of an element have the same Z (number of protons or electrons) but different numbers of neutrons and hence different A (mass number). Chemically, all isotopes of the same element are identical. el Raphex 2007 3General ~ Answers GI6. C Tritium has one proton, two neutrons, and one electron; hydrogen has one proton and one electron. GI7. C Gis. E The mass number (A), 59, is the number of protons (Z) plus the number of neutrons (N). The atomic number (Z), 27, in an electrically neutral atom, is equal to the number of electrons. GI9. E Only a single photon can ionize an atom. If the energy of one photon is insufficient for ionization, then ionization does not occur. G20. D The binding energy of the L-shell is usually significantly lower than for the K-shell. G2l. D Binding energy increases slightly after de-excitation to the ground state via gamma-ray emission. G22. B G23. A N is the number of remaining radioactive atoms. The “~—” sign corresponds to the fact that number N is decreasing with time. G24. C After a half of T,2, the initial activity of the isotope A decreases to A*27!? = 0.707 A. G25. D The decay constant d is equal to 0.01/day. T 1. = 0.693/A = 69.3 days. G26. C 1/Teg = [1/T, + 1/T)). G27. A An alpha particle consists of two protons and two neutrons. G28. A A beta-plus particle is a positron that has electric charge of +1. 4 Raphex 2007General ¢~ Answers “ G29. D Positron emission and electron capture often compete as a decay process in the same radionuclide with an excess of protons. In electron capture, an electron, usually from the K-shell, combines with a proton to form a neutron and an emitted neutrino. The K-shell vacancy causes characteristic x-rays and Auger electrons to be emitted. G30. B \8F is a positron emitter used in PET. '°7Cs is a reactor fuel by-product, created by fission. \°2[r is created by bombarding '9'Ir with neutrons in a reactor. “°K is a naturally occurring isotope of potassium, present in the human body. G3l. B G32. B G33. C 1 Ci = 3.7x10"° Bq. G34. B G35. D Exposure rate = (Exp rate const.) x A/? = 3.3 R.cm?.mCi'.hr! x 163 mCi x 1/300)? cm? = 6 x 10-3 R/hr. G36. C Exposure rate at 30 cm = Exp. At 1m x (100/30)? = 1.75 x 11.11 = 19.4 R/hr with no shielding. G37. D The transformer is used to create high voltage across the tube. G38. B Electrons travel from the cathode to the anode. G39. D X-ray production is a very inefficient process, and steps must be taken to dissipate the large amount of heat generated. G40. B G4l. C G42. D SS Raphex 2007 5General «<> Answers & G43. B Possible characteristic x-rays energies are found by taking the energy differences between electron shells. Bremsstrahlung (“braking radiation”), which is energy lost by the electron in the field of the nucleus, can have any energy up to that of the incident electron. G44. C The maximum bremsstrahlung energy in keV is equal to the maximum energy of the incident electrons, i.e., to the kVp applied across the tube. G45. A The minimum photon energy depends on the filtration. G46. A This is due to beam hardening as the beam passes through the first HVL. G47. C G48. D As photons carry zero electric charge, they are not deflected by a magnetic field like charged particles. G49. D Gamma rays originate from nuclear decay, while x-rays are produced by electrons slowing down in the field of the nucleus, or following electron transitions between shells. G50. B GSI. B E = hv = he/A, where h is Planck’s constant, c is the speed of light, and A is the wavelength. G52. D The formula for the intensity of radiation at a depth x is: I, = Ip eM = Ip e- O8WHVL, where Jy is the initial intensity. G53. Bo = Ly cP = Ip er OSXHVL 2 Jp @0593"32 = 0.354 Ip, G54. B G55. A /=1p. G56. D 6 Raphex 2007General < Answers G57. D Increasing the tube current or mAs increases the intensity but does not change the spectrum. Decreasing the filtration actually decreases the effective photon energy. G58. E Increasing the effective energy by additional filtration increases the penetration and thus reduces the skin dose for the same intensity of the exit beam. G59. B Coherent scattering does not change the beam energy. The pair production threshold is above the energy of a typical diagnostic x-ray unit. The Compton effect may slightly increase the effective beam energy but not nearly to the extent of the photoelectric effect which has Z?/E? dependence of the interaction probability. G60. D The probability of the photoelectric effect per unit mass is proportional to Z?. Gél. D The photon is fully absorbed in a photoelectric interaction. G62. A There is no energy loss in coherent scatter. G63. D The probability of Compton interactions decreases with increasing photon energy. G64. B The Compton photon can be scattered at any angle, but the emitted electron is limited to 0°-90° with the direction of the incident photon. G65. C G66. C The threshold of pair production is 1.02 MeV. G67. A From the initial 6 MeV, 1.02 MeV is used to create a positron and an electron, and the remaining 4.98 MeV is divided between the particles as kinetic energy. After losing its kinetic energy, the positron will undergo annihilation with an electron at rest, emitting two annihilation photons each of energy 0.51 MeV. G68. D Compton interactions are most probable in soft tissue between 25 keV and 25 MeV. G69. B See answer to question G67. er LS Raphex 2007 7General ~ Answers G70. A Betas travel about 0.5 cm per MeV in unit density material. In air, they will travel about 1/0.0013 or 800 times as far. The photoelectric interaction is a photon interaction with matter. After a beta decay of a neutron, a proton, an electron, and an antineutrino appear, so the total electric charge is conserved. Grids tend to increase patient dose (for the same optical density on the film), but they intercept scatter generated by interactions in the patient, which would reduce contrast. Optical density is the logarithm of the ratio of the incident light intensity divided by the transmitted intensity. Thus optical density is additive. 68% of the measurements fall within +o of the mean, where o is the standard deviation equal to N'. N is the total number of counts. Therefore, o = (1000)!? = 32 counts. G7l. C G72. C G73. C G74. A G75. B G76. B G77. B G78. C G79. A G80. E G8I. D G82. C 8 Raphex 2007General + Answers ¢ G83. D Although most studies on cancer induction by radiation have been based on doses from therapeutic x-rays (and on data from bomb victims), repeated fluoroscopy of female TB patients using now-obsolete equipment has shown an increased incidence of breast cancer in this population. (Note: These patients received relatively high exposures; this should not be taken to imply that routine, low-dose mammography would have the same effect on the population.) The absence of data showing an increased incidence in other sites may only mean that the doses and numbers of cases are too small to be statistically significant. G84. D G85. B In NCRP Report 116 the Effective Dose (E) attempts to weight the radiation dose to different organs by the relative cancer risk of each organ. G86. C The highest risk from 2 to 16 weeks post-conception is a maximum incidence of 1% per rem for small head size. G87. A According to the NRC, the radiation badge must be worn on the front of the trunk between the head and waist. The badge should face away from the body. G88. C The recommended effective whole-body dose limit for radiation workers is 50 mSv/yr, or 1.0 mSv/wk. G89. E The instantaneous dose rate is not a factor in barrier thickness calculations. However, the total dose in a week must be kept below the recommended limit for workers or the general public, depending on who will occupy the area. G90. A B and C result in lower and higher charge collection, respectively. The temperature- pressure correction gives the reading that would have been measured with the chamber at standard temperature and pressure. G9l. E G92. B Raphex 2007 9General + Answers % G93. D G94. D G95. D G96. C The advantage of a Geiger counter is its sensitivity. The ionization by a single event is magnified, due to gas multiplication, into a measurable signal. Its disadvantage is low accuracy and saturation or suppression of the signal in a high-intensity field. When an unshielded container does not give any reading above background as measured with a sensitive detector, the contents may be disposed of as non-radioactive material. Once installed, x-ray units are regulated by the state. Mammography units are also regulated by the FDA, under MQSA standards. The FDA regulates the manufacture and installation of x-ray devices, under 21 CFR 1020. The NRC regulates the use of “man-made” radioactive materials such as Co-60 units and brachytherapy sources. ST Raphex 2007Diagnostic ~ Answers DI. B The filament current determines the filament temperature and thus the rate of thermionic electron emission. The space charge cloud shields the electric field for tube voltages below 40 kVp, placing an upper limit on the tube current. This effect is overcome by applying higher kVps. The limiting factor in using higher mA is the heat generated in the anode of the tube. Mammography tube current is typically 100 mA. Due to the extended exposure times in fluoroscopy, the fluoro tube currents (about 1-4 mA) are much lower than those used in radiography (about 50-800 mA). D2. C The power rating of x-ray generators (kW) is the average power delivered by the maximum tube current (An,x) for 100 kVp and 0.1-second exposure time. D3. C Small anode angles (7° to 9°) are in fact desirable for small field-of-view image receptors, such as neuroangiographic or cineangiographic equipment, where the field coverage is limited by the image receptor diameter (e.g., 23 cm). This allows smaller effective focal spot size. D4. A Spectrum II has more higher-energy photons, hence a higher effective beam energy. Ds. B Characteristic radiation from both spectra appears at the same energy. This is only possible when the same target material is used. D6 A The minimum wavelength occurs at the highest beam energy, which is the same in both spectra. D7. B Differences between the two spectra’s effective beam energies and total radiation output (area under the curve) indicate that the spectra are produced using different phase generators. bs. A The effective focal spot size changes direction along the anode-cathode axis. Projections of the focal spot on the anode side of the field cause a decrease in the effective anode angle, resulting in a smaller projected focal spot according to the line focus principle. Therefore, the focal spot becomes larger on the cathode side, with less resolvability. No changes occur in the perpendicular direction. DY «6©CE Phototimers operate by measuring the actual amount of radiation incident on the image receptor and terminating x-ray production when a preselected reference radiation exposure value is reached. In film-screen radiography, a user-selectable “density” switch on the generator control panel adjusts the reference value. Lowering the “density” setting from 0 to -2 will result in reduced exposure at the screen-film system causing overall reduced film density. Raphex 2007 WlDiagnostic “ Answers “ DIO. E Non-stochastic biologic effects occur at doses above 1 Gy. The entrance dose from a screen-film AP lumbar spine is less than 5 mGy, while the CT of the abdomen has a skin surface dose of about 20 mGy. The phototimer maintains the same exposure to the detector, resulting in a lower dose to the smaller patient. The dose to the patient is inversely proportional to the speed of the screen-film system. The dose is higher near the entrance point and lower near the exit point of the patient. 30% is absorbed in the first screen, 70% passes through. The second screen absorbs 30% of that 70% (or 21%). Total stopped is 30% + 21%. OD = logio(I/T), where I and T are the incident and transmitted light intensities. For OD = 2.0, T = 0.01 (1%), and for OD = 1.0, T = 0.1 (10%). Optical densities are additive, so the net OD of both films is 3.0. An OD of 2.0 looks dark on a standard view box. By similar triangle geometry, the penumbra is equal to the focal spot size, since source- object and object-film distances are equal. Screen quantum detection efficiency is higher for thick screens, which contain more absorbing fluorescent material. Detail screens are thin screens associated with less light diffusion in the body of the screen, hence less blur and better resolution. Cesium iodide is an excellent scintillator used in fluoroscopy and digital radiography; however it is too sensitive to moisture to be used for screen-film radiography. This is due to increased attenuation coefficients with reduced kVp; A and E will reduce subject contrast. Focal spot size affects resolution, not contrast. An upside-down grid will produce a film that is dark in the center and almost blank toward the edges. Patient motion produces a blurred image; lateral misalignment produces an over- all lighter image; poor screen-film contact produces blurring on the image. The HVL for screen-film mammography in mm Al should be greater than the kVp/100, but less than the kVp/100 + 0.1. For example, at 27 kVp it should be between 0.27 and 0.37 mm Al. A Mo filter is commonly used with Mo anode mammography x-ray tubes, but HVLs are always measured in mm Al. DIT. C DI2, D DI3. C DI4, C DIS. B DI6é. A DI7. A DI8. C 12 Raphex 2007Diagnostic “ Answers’ DI9. D20. D2I. D22. D23. D24. D25. D26. Raphex 2007 The focal spots for fluoroscopy are typically 0.3 or 0.6 mm; those for standard radiography are usually 1.0 to 1.2 mm. The spatial resolution for fluoroscopy is usually limited by the TV system to 1.8 to 2.5 Ip/mm, while radiography has resolutions of 4 to 8 Ip/mm. The tube current for fluoroscopy is usually 1 to 3 mA in order to limit anode heating for the long exposure times of 3 to 10 minutes; because of the short exposure times (less than 1 second) of radiography, tube currents of 200 to 800 mA can be used. Tube potentials are similar for both procedures. SSDs in fluoroscopy are typically 18, while the SSDs for radiographs are typically about 25 inches (except for chest radiographs). Penumbra, caused by a finite focal spot, increases with magnification. Eventually, this dominates the image. The grid, H&D curve, and size have no effect on magnification. The receptor’s MTF becomes less important as magnification increases. A high base+fog (normal is <0.25) can be from contaminated developer, high developer temperature, or film fogged during storage or handling. The Compton scattering mass attenuation coefficient decreases with energy. Over the diagnostic range, however, it changes little, and may be viewed as constant. This term represents the ability of a system to reproduce an object whose linear attenuation coefficient does not vary greatly. From the line-focus principle, the effective focal spot will be smaller at the anode-side of the x-ray field. Therefore, the limiting resolution should be somewhat greater at the anode side (but not doubled). The fluoro kVp for small patients will be low, in which case the Compton scatter will be low. Removing the grid means more x-rays reaching the II, so patient exposure will decrease, which is especially good for infants whose risk from ionizing radiation is higher. Scatter does decrease with decreasing thickness and field size. Since the II cannot be brought all the way down the table, there will be a large air gap, which is effective at removing scatter. Well-hardened means we can assume monochromatic-like exponential attenuation, and thus attenuation of about 20% per cm. The transmission through the extra 5 cm would be 0.8 x 0.8 x 0.8 x 0.8 x 0.8 = 0.33. Thus an increase in the patient entrance exposure of about 3x is necessary to maintain proper film exposure.Diagnostic ~ Answers D27. E The heel effect causes a lowered film density toward the anode side of the film that can be dramatic near the edge of large films at shorter SID if the anode angle is small enough. High ratio parallel grids produce film density which decreases away from the center due to grid cut-off. Inverted grids have density only in the central region. Worn screens produce spots of low density in the worn regions. Phototimer drifts uniformly increase or decrease the film density. D28. B B will generally require a decrease in current, which will require a longer exposure time. A, C, D will allow a reduction in exposure time. D29. B Geometric unsharpness = (focal spot size) x (magnification — 1). When the x-ray tube of the C-arm is closer to the patient, the magnification increases, causing more unsharpness. Three-phase generators allow increased mA and increased x-ray output per mA, resulting in shorter exposure times and less motion blurring. Contact radiography with no magnifica- tion means the screen determines the image sharpness. A single thin screen is sharper. D30. A CTDI,,; is just CTDI9/pitch. CTDI,; is typically 40-60 mGy for an adult cerebrum but doubles for a study with and without contrast. Typical values are: 10-35 mGy for an adult abdomen, 5-10 mGy for an adult chest, 60 mGy for most coronary CTA studies. Pediatric studies should always have a lower dose than adult studies (i.e., lower mAs and/or kVp). The brain requires the best SNR to see tumors and other abnormalities, and thus it requires the highest radiation dose. D3I. D A head is low (4-6 mSv) because of low tissue sensitivity but doubles for studies with and without contrast. An abdomen is 5~10 mSv. A chest is somewhat lower than an abdomen because the mAs is lower. A coronary CTA has a high mAs leading to a higher dose (~60 mGy) and high tissue sensitivity for an effective dose of 10-15 mSv. Pediatric dose should always be lower because of lower mAs and/or kVp. D32. E For 3-D reconstructions, the best image will be from the thinnest slice with the smallest slice increment to avoid discontinuities (banding, stairstepping) in the 3-D image. D33. A Hounsfield number = 1000(Umateriat — Uwater)/Hwater: By definition, Hounsfield number is a ratio and has no units. Window variations, matrix size, and mAs have no effect on the tis- sue CT number. D34. D Measured slice width is always greater than nominal for 5 mm slices, and helical slice width is greater than axial for pitch 1 due to the interpolation algorithm, which averages data from more than one collimation width. \4 Raphex 2007Diagnostic ~ Answers D35. D36. D37. D38. D39. D40. D4l. D42. D43. p44. Although the patient dose is lower for a pitch greater than 1, it will be higher for a pitch of less than 1, since this represents an overlapped x-ray beam on the patient surface. The reconstruction algorithm affects both noise and resolution, but has no effect on dose. A smooth algorithm decreases noise and resolution. A sharp algorithm increases noise and resolution. Chest units have relatively low workload because of low mAs/image. Mammography, even at the higher kVps used in FFDM is still low kVp, and hence low penetration. CT has high penetration because of the high kVp and a very high workload because of the high mA (400-500) and long scan times. Mobile C-arm fiuoroscopes are generally used in procedures with a low beam-on time per day, and hence have a low workload. FFDM CR mammography is performed with standard mammography x-ray units. The probability that a “called abnormal” in screening is a cancer (PPV1) should be between 5% to 10%. A higher PPV means the radiologist is calling too few abnormals. A lower PPV means overcailing. Poor image quality would not generally lead to overcalls, but it could lead to undercalls. This order results in the minimum light diffusion in the screen-film system and thus the best resolution. Rh filter produces a more penetrating x-ray spectrum than Mo (higher HVL) and thus will reduce dose, but with an accompanying decrease in contrast. All these items can cause artifacts on the film, but white specks are due to either dust on the screen or processor roller pick-off. 4300x4300. The matrix size is FOV/pixel size = 430 mm/0.1 mm = 4300. Since it is a square FOV, the matrix size will be 4300x4300. 5 Ip/mm. 100 micron (or 0.1 mm) is the smallest object resolvable without aliasing. Since a line pair (Ip) consist of two objects, the line and the space between them, two detector elements are needed to image the smallest possible lp. This requires 2 x 0.1 = 0.2 mm. Thus Ip/mm = 1/0.2 = 5 Ip/mm. This is the definition of the Nyquist frequency. In general, Ip/mm = 1/ 2d, where d is the smallest object resolvable. Raphex 2007Diagnostic + Answers D45. D46. D47. D48. D49. DSO. DSI. A CsI has a high absorption efficiency and conversion efficiency, and is used as the fluores- cent phosphor in several DR designs. Nal is used in gamma cameras; BaFBr/BaFI mixture is the storage phosphor used in CR; GdOS is used as a screen-film phosphor and as a CT detector; and YGdO is used as a CT detector. A thin phosphor must be used to maintain a reasonable resolution (~5 Ip/mm) since CR is usually a single-screen system with the light read out from one side. If a thicker phosphor were used to increase the absorption efficiency, the resolution would degrade because of light diffusion. Dual-screen, dual-side-reader CR is becoming available to overcome this limitation. Screen-film systems use dual phosphor. The number of shades of gray = 2*i8 = 2!9 = 1024. Bits are grouped into bytes, each consisting of 8 bits. 10 bits will require 2 bytes of storage. Digital detectors are inherently signal-to-noise ratio (SNR) limited, because of the ability to vary the image display characteristics by post-processing (contrast enhancement, spatial resolution enhancement, etc.). The mAs is linearly related to the number of x-ray photons incident on the patient and also to the number of x-ray photons absorbed by the x-ray detector. Therefore, the SNR (N/ VN =\N) is determined by the number of x-rays detected and converted to an output signal. Amorphous selenium is a direct detector used in flat-panel systems. The electrons released in the detector from x-ray interactions are used to form the image directly. When x-rays are absorbed by photostimulable phosphors used in CR, some light is prompt- ly emitted, but much of the incident energy will be stored in the phosphor, later released in the form of light using a readout laser beam. Cesium-iodine crystals emit light promptly when struck by an x-ray beam. Charge-coupled devices (CCDs) form images from visible light. They are made of crystaline silicone, and are commonly used in fluoroscopy and digital cineradiography. There must be a reduction by a factor of four in maximum frame rate because there will be four times as much data. The data transfer rate in the system does not change. Doubling the matrix size can double the resolution if the other components are adequate. It is similar to doubling film or screen resolution. X-ray exposure per pixel must be maintained in order to maintain quantum mottle. Since there are four times (1024/512) as many pixels, the exposure must be increased by a factor of 4. Reducing the effective field size reduces the brightness gain by reducing the minification gain. The ABC system compensates by increasing the exposure factors, increasing patient entrance exposure. Raphex 2007Diagnostic + Answers’ D52. DS53. DS54. DS55. D56. D57. DS58. D59. Edge packing is the high count rate found around the outer edge of a scintillation camera. Scatter is most intense where the X-ray beam enters the patient. Scatter on the exit side is typically 10% of the intensity on the entrance side. The limiting entrance skin expsoure (ESE) rate of 10 R/min is about 0.1 Gy/min. As a rule of thumb, the exposure rate duescatter radiation at a distance of 1 meter from the patient’s center is about 1/1000th of the ESE rate. Knowing either fluoro time or DSA frames alone gives insufficient information to predict the possibility of deterministic injury. Both fluoroscopic and DSA dose rates are influenced by patient size. The risk of deterministic injury increases if the beam is not moved during the procedure. Vertical resolution = (Kell factor x #lines)/(2 x FoV) in mm. True events refer to the coincident detection of the two opposed 511-keV gamma rays resulting in a line of response passing through the site of the decay. Randoms are when two 511-keV gamma rays are detected from two separate positron decays. These events contain no positional information from the distribution of radioactivity in the body. Scatter events are when one or both photons emanating from the same decay site undergo small angle Compton scattering. Because the photons lose only a small amount of energy, their energy remains in the photopeak and is not eliminated by energy discrimination. The definition of a mis-administration in a nuclear medicine diagnostic exam is a discrep- ancy in the administered activity of >50% of the intended dose. A discrepancy of >10% applies to radionuclide therapies only. Radiochemical purity refers to the fraction of radioactivity associated with the desired product. It is determined by chromatography and is the fraction of the total counts in the peak corresponding to the desired radiopharmaceutical (the rest being usually metabolites or unconjugated radionuclide that was not removed by purification). It is usually required to be greater than 90 or 95%. Poisson statistics are the statistics that govern random events for which the probably of any specific event occurring is small. It expresses the probability of a number of events occurring in a fixed period of time if these events occur with a known average rate, and are independent of the time since the last event. This is the situation for radioactive decay. The Poisson distribution is a special case of the binomial and Gaussian distribution, namely when p, the probability, is small and the number of events, N, is large. Raphex 2007Diagnostic + Answers D60. C The number of radioactive atoms, N, is given by the formula for activity A = AN where A is the decay constant given by the formula 4 = 0.693/t,. where t),) if the isotope half-life. Rearranging these equations: N = 1.44 x ty; A = 1.44 x 110 min x 60s/min x 10 x (3.7 x 10’) (Bq/mCi) = 3.52 x 10”. Dél. E Ideally, the image corresponding to points on the central axis of a SPECT camera should remain constant as the camera rotates around the patient. Unfortunately, due to the weight of the cameras (head sag), the mechanics, the effect of magnetic field of the earth on the PM tubes, etc., imperfections exist between the correspondence of points in the image acquired at different angles relative to the locations in space to which they correspond. These imperfections are greatly reduced by the application of a COR correction, which is applied at each angle in a SPECT image acquisition. Without a COR correction, the SPECT-constructed image would be blurry and would contain ring artifacts. D62. D A well-scintillation counter is highly sensitive and able to count radioactivity within a range from the minimum limits of detectability (approx. 3 times background) up to about 1 microcurie. Most automated well-counters possess energy discrimination and can count isotopes in pre-determined energy windows. Clinical dose calibrators are too insensitive to measures wipes and have no energy discrimination. Radiation survey meters have adequate sensitivity but cannot discriminate readily between different isotope energies. A gamma camera would be too insensitive to measure the spectrum. D63. C For the typical cruising altitudes of commerical airline flights, the reduced atmospheric “shielding” of cosmic radiation from outer space results in higher radiation dose rates than those at sea level, by approximately 0.5 mrem/hr. Dé4. B Many radiopharmaceuticals are secreted into breast milk and will therefore be ingested by a breast-feeding infant. Given the relatively large activities thus ingested and the small organ masses of an infant, the absorbed doses to a baby may be considerable—up to tens of rads or even higher, depending on the radiopharmaceutical and activity administered to the mother. Accordingly, breast-feeding should be discontinued until the activity in the mother has decreased (through physical decay and biological elimination) to negligibly low levels. For virtually all °™Tc-labeled radiopharmaceuticals, this will be about 1 to 2 days following administration of the radiopharmaceutical. For longer-lived ©’Ga-labeled agents, on the other hand, this may be as long as 1 to 2 weeks. For '*!I, breast-feeding should be discontinued permanently for the current baby. D65. C The transport index (TI) of a radioactive package is defined as the exposure rate in mR/hr (milliroentgen per hour) measured at 1 meter from the surface of the package. For a pack- age yielding an exposure rate of 0.002 R/hr, or 2 mB/hr, at 1 m, the transport index is therefore 2. 18 Raphex 2007Diagnostic “+ Answers Dé6. B Dé7. B D68. A D69. C D70. A D7l. E D72. D The limited patient data available on reproductive function (fertility) following high-dose radioiodine therapy are suggestive of a slow (up to several years), but complete, recovery of such function. While fertility rates appear to be reduced within the first year post- treatment, they appear to be similar to historical averages by approximately 10 years post-treatment. Uranium is a naturally occurring component of the earth’s crust throughout the world. As it decays, it produces gaseous radon, which can diffuse into the air. As we breathe, we unavoidably inhale and exhale this air-borne radon. However, some of this inhaled radon may spontaneously decay to polonium within our lungs before it is exhaled. Polonium is a solid, not a gas. Therefore, intrapulmonary polonium atoms will settle into the surface of the airways. When this deposited polonium decays, a densely ionizing, high-LET alpha-ray track passes through the lung tissue. Because of the high quality factor of alpha rays (20) a large dose equivalent is delivered to the lungs; this results in a large contribution to the background effective dose equivalent. Historically, the Nuclear Regulatory Commission (NRC) and Agreement States required patients receiving radionuclide therapy to remain hospitalized until the retained activity in the patient was less than 1110 MBq (30 mCi) or the dose rate at 1 m from the patient was less than 0.05 mSv/hr (5 mrem/hr). However, in 1997 the NRC amended its regulations concerning radionuclide therapy patients through the issuance of new rules that appeared in the Federal Register on January 29, 1997. The current NRC regulations, revised 10 CFR 35.75 effective May 1997, allow for the release from medical confinement of patients if the expected total effective dose equivalent (TEDE) to individuals exposed to the patient is not likely to exceed 0.5 rem (500 mrem). The quality factor (QF), now also known as the radiation weighting factor (wr), reflects differences among radiations in producing biological damage, as does the relative biological effectiveness (RBE). However, the RBE is impractical to use for radiation protection purposes as the actual value of the RBE depends on the energy of the radiation and (quantitative) biological effect being considered. The quality factor, on the other hand, is essentially a “typical” value of the RBE assigned to a particular radiation. Radioactive waste being held for decay in storage can be disposed of as non-radioactive waste only after it has been verified that the waste is no longer detectably radioactive. Assay of such waste should be performed using the most sensitive type of survey meter, namely, a solid-state (crystal) survey meter, and not a much-lower-sensitivity Geiger counter. Raphex 2007Diagnostic “* Answers D73. D D74. A By all relevant criteria—the large size of and the high number of excess cancers and other D75. D76. D77. D78. Cc D D Cc diseases in the exposed population, a large, well-matched control (i.e., unexposed popula- tion), remarkably accurate and precise radiation dosimetry (with a broad radiation dose range), and excellent, long-term medical follow-up—the Japanese-American Radiation Effects Research Foundation (RERF) follow-up study of the Japanese A-bomb survivors remains the most fruitful and most reliable study of radiation effects in man. With the publication of the BEIR VI and BEIR VII Reports, the linear, non-threshold model is now the prevailing model. It implies that the risk per rad is constant at all doses, and it further implies that any excess radiation dose (i.¢., above background), no matter how small, carries with it a finite excess risk of cancer and genetic damage. The available human data are also consistent with the linear, non-threshold model; that is, on the basis of statistical considerations such as goodness of fit, one cannot actually choose between the e sub-linear (i.e., linear-quadratic) and the linear, non-threshold models. The linear, non- threshold model is more conservative (i.e., predicts a higher risk per rad at low doses) than the sub-linear model and is therefore considered “safer” and thus more appropriate by some for radiation protection purposes. Largely on this basis, it has been adopted by the BEIR Committee and others. As noted, however, this remains highly controversial and unresolved. Signal-to-noise ratio (SNR) is linearly proportional to the voxel volume, which increases with slice thickness. The SNR is proportional to the square root of the number of signal averages (number of acquisitions). The proximity of the receiver coil to the volume of interest increases the coil quality factor (which increases the SNR). Body receiver coils have moderate quality factor. A narrow bandwidth (a narrow spread of frequencies around the center frequency) provides a higher SNR that is inversely proportional to the square root of the bandwidth. Chemical shift refers to the resonance frequency variations resulting from intrinsic magnet- ic shielding of anatomic structures. Resonance frequencies for fat protons are lower than for water protons. Such a frequency shift (about 3.5 Hz/MHz of central frequency) will cause a shift in anatomy (misregistration of water and fat moieties) along the frequency encode direction. A higher magnetic field strength increases the central frequency, and will subsequently increase the fat-water frequency shift. When using lower gradient strengths, water and fat are not contained in the same pixel boundary but separated by one or more pixels. Swapping the phase and frequency encode gradient directions can displace (not eliminate) chemical shift artifacts from a specific image region. The proximity of the RF receiver coil to the volume of interest (patient in the gantry) increases the coil quality factor (which increases the SNR). 20 Raphex 2007Diagnostic “+ Answers’ D79. B Ds0. C Dsl. A D82. B D83. D D84. B D85. C D8. B D87. C Single or multiple Echo Planar Imaging (EP!) acquisition is a technique that provides extremely fast imaging time. EP images typically have poor SNR, low resolution (64x64 or 64x128 matrix) and chemical shift artifacts. Nevertheless, EPI offers real-time “snap- shot”image capability with 50 msec or less total imaging time. EPI is most useful for time- dependent physiological processes and functional imaging. Inversion Recovery acquisition technique emphasizes the T1 relaxation time of the tissue by extending the longitudinal recovery by a factor of 2 with an initial 180° RF inversion pulse. Variations in the time of inversion, TI, controls the contrast between images. Short Tau Inversion Recovery (STIR) uses very short TI. During longitudinal recovery, all tissues pass through a bounce point or tissue null. Judicious TI selection can supress a given tissue or signal such as fat or chemical shift artifacts. The use of longer TI as in Fluid Attenuated Inversion Recovery (FLAIR) reduces the signal levels of tissues with long T1 relaxation constants, like CSF and other water-bound anatomy. MR Perfusion and Diffusion imaging use endogenous or exogenous “tagged” tracers to study the delivery of oxygen and nutrients to the cells and removal of carbon dioxide from the cells. Blood oxygen level-dependent (BOLD) and “functional MR” imaging rely on the differen- tial contrast generated by blood metabolism in active areas of the brain. Electronic instruments that depend on electron focusing, etc., are very sensitive to changes of even the earth’s magnetic field. Therefore, fringe fields greater than 1 gauss can cause severe problems. The safety limit is 5 gauss (pacemakers, etc.). The earth’s magnetic field is about 0.5 gauss. For a typical gradient strength of 0.5 g/cm, 0.5 g/cm x 30 cm = 15 gauss. Gradient field strength ranges from about 0.1 to 1 gauss/cm. The T2 relaxation time only changes slightly with field strength at fields <4T. To a first approximation, one can assume it is unchanged. For a 4-MHz transducer, the attenuation is 4 dB/cm x 7.5 cm = 30 dB. The reflection coefficient R = (Z2 - Z1)? / (Z2 + Z1)* between two tissues with acoustic impedances Z1 and Z2 is unaffected by an interchange of Z1 and Z2. Raphex 2007 21Diagnostic “+ Answers” D89. D90. D9. D92. D93. D94. A Axial resolution = '/, spatial pulse length. The pulse length = wave length x number of waves in a pulse = 0.33 mm x 3 = | mm. Axial resolution = 1/, x 1 mm = 0.5 mm. Echoes from surfaces further apart than that will not overlap, and can be resolved. Q factor = operating frequency/bandwidth. The Q factor increases with increased pulse length due to narrow bandwidth. A decreased (short) pulse length has a much broader bandwidth and therefore decreased Q factor. Lateral resolution is independent of pulse length. The beam width and number of scan lines are the major factors determining the lateral resolution in the image. A decreased pulse length will improve the axial resolution. The acoustic impedance of tissue depends only on the density and the speed of sound. The speed of sound in tissue depends on the tissue, and is independent of the pulse length. Ultrasound contrast resolution depends on several interrelated factors. Acoustic impedance differences give rise to reflections that delineate tissue boundaries and internal architec- tures. The density and size of scatterers within tissues or organs produce specific “texture”. With proper signal processing, attenuation differences result in gray-scale differences among the tissues. Atomic number differences do not produce acoustic contrast. A mirror image artifact arises from multiple beam reflections between a mass and a strong reflector, such as the diaphragm. Reverberation artifacts occur between two strong reflectors producing equally spaced signals of diminishing amplitude in the image. Shadowing occurs distal to objects of high attenuation, resulting in hypointense signal. Speed of sound variation in the tissues can cause mismapping of anatomy. Side-lobes energy emissions in transducer arrays can cause anatomy outside of the main beam to be mapped into the main beam. In pulsed Doppler the maximum blood velocity, Vina, that is accurately determined is increased with larger PRE, lower operating frequency (longer operating wavelength), and larger Doppler angle. Vinax = C x PRF / 4 x fy x cos(8). C is the speed of sound in soft tissue. A 95% confidence interval means the counts must fall within two standard deviations (SD) of the mean (N). Error limit = 1% = 2 SD/N, but SD = N!?. Thus 0.01 = 2(N!?)/N = 2/N!?. [0.01] = 4/N. N = 40,000. 22 Raphex 2007Diagnostic “- Answers DIS. E The relative standard deviation (RSD) of the net counts of a sample is given by: RSD = (G + B)!2 / (G - B) where G = gross counts and B = background counts. Thus RSD = (122 + 22)!” / (122 - 22) = 0.12. D996. C 1/16 inch (about 1.5 mm) of lead is used. The equivalent thickness of concrete would be 10 to 15 cm. Aluminum is not generally used, but a much greater thickness would be required. D97. C Db98. A D999. D D100. C In general radiography rooms, a wall intercepting the primary beam (a wall holding the bucky and/or the floor) is called a primary wall. In CT scanner, fluoroscopy, and mammogra- phy rooms, all walls in the room are secondary barriers. Primary barriers in such modalities are provided by the CT detectors and the fluoroscopy image intensifier, and in mammography by the receiver and receiver holder. Raphex 2007 23Therapy “+ Answers + Th OD Tm. 2B T3. D T4. = #B Ts. OA TT. C Approximate dose at isocenter from AP image (assuming 1 MU = 1 cGy) = MU x TMR x OF x #fractions = 3 x 0.752 x 1.071 x 42 = 101.5 cGy (assuming a 15 x 15 cm field). The dose from kV images is negligible at the isocenter. For linac-based kV imaging systems mounted at 90° from the MV beam axis, one can either take orthogonal kV images, requiring gantry rotation between the images, or one MV and one kV image, which requires no gantry rotation. In the latter case, although the technique may save time, and deliver acceptable image quality, the dose from daily MV images may be a considera- tion: the field is always in the same direction, may be larger than the treatment field, and the dose will be even higher if images are repeated to verify a shift. T7. A Equivalent square for 10 x 20 is 13.3 cm. MU =~ -aamsonny OFx me (1.023%0.874) 100 Ts. OD The maximum dose is the dose at depth d,,,, = 250 Xx (1.0/0.874) = 286. TT OD Dose at dina, expressed as a percent of dose at midplane for parallel-opposed fields, decreases with increasing energy and increasing SSD. TIo. B PDDs increase with increasing SSD, because the inverse square component decreases. The correction (known as Maynard’s F factor) in this case is: 2 2 PDD (150 SSD) = PDD(100 Ss) x| HO] x | (50+ nae) (100+d,,,,) 160 Dynax is a small factor, and can be ignored. Thus, the factor is 1.063, i.e., a 6.3% increase. 24 Raphex 2007Therapy + Answers TH. TI2. TI3. T14. TIS. TI6. TI7. T18, TI9. 720. T2I1. The equivalent square of a rectangular field has approximately the same PDD and TMR as the rectangle. It is smaller in area than the rectangle (i.e., C x C < 6 x 30 in this case), since it is the field with the same scatter contribution on the beam axis. A useful rule of thumb is that C = 4x (area/perimeter). The use of “equivalent square” enables PDD and TMR tables to be simplified to only square fields rather than tabulating the many rectangu- lar fields in use. MU = 60/(0.90 x 0.782 x 0.58) = 147 MU. For opposed tangents, the wedge angle affects dose homogeneity in the axial plane only. Techniques A, B, or C would be needed to decrease the hot spot inferiorly. All the other wedges are reversed. The surface dose is about 15% to 40%, depending on field size. Bolus can be used in an electron beam both to increase the surface dose (although this is usually fairly close to the prescribed dose) and to adjust the depth of the prescription isodose curve in part or all of the beam. A 1-cm Lucite™ spoiler can be placed in the beam, close to the patient. IMRT can be performed with custom compensators in each beam. Although this approach is generally slower than IMRT with MLC, it does allow IMRT to be made available on older machines for which retrofitting an MLC would be too costly. Raphex 2007 25Therapy + Answers ¢ 722. T23. 724. 725. 726. 727. T28. T29. T30. T3!. 732. For example, in a mantle field, a more homogeneous dose distribution can be achieved if the areas of “missing tissue” are replaced with an equivalent thickness of compensator material, demagnified to the projected size on the blocking tray. Commercial milling machines that make Cerrobend® forms can be used to fabricate custom compensators. The level of dose homogeneity required clinically must be evaluated by the physician, since missing tissue compensators could be made for every treatment field, but are required in relatively few cases. Compensation with wedges is often adequate. By similar triangle geometry: gap = d x (C/2)/SAD for each beam = (9 x 10/100) + (9 x 12/100) = 2.0 cm. The angle of divergence is the angle whose tangent is the beam half-width/100; ie., tan“! (7.5/100). This is important when angling beams to remove divergence (e.g., at the chest wall for opposed tangents). Answer D will match the dose on the skin, but create a cold triangle at depth. Higher photon energy is generally an advantage when treating a large thickness of tissue because of the greater dose uniformity and lower skin dose. However, it can be a disadvan- tage when the dose in the build-up region is too low; for example, when treating superficial nodes. For the same reason, high-energy beams can cause a lower dose at the lung-tissue interface, possibly underdosing part of the treatment volume. The effective TMR will not change, and the collimator output (Sc) will change minimally, so this is essentially an inverse square problem. MU at 140 SSD,d5 = 320 x (145/125)? = 431 MU. 26 Raphex 2007Therapy + Answers 733. 734, T35. T36. 137. 738. T39. T40. T4i. T42. Ignoring lung corrections tends to overdose the prescription point due to the lower attenua- tion of lung compared with muscle tissue. Thus it is important to adjust the prescribed dose depending on whether or not lung corrections are used. The higher the photon energy, the less the correction. However, it has been suggested that photons above 12 MV should not be used because of the build-up effect as the beam passes from lung into unit density tis- sue, possibly underdosing tissue in this region. The tissue in contact with the metal rod will receive significant scatter dose from the metal. This process is generally not well modeled by current commercial treatment planning systems. The dose “downstream” from the metal rod (beyond the distance where interface effects are important) should have acceptable accuracy regardless of the correction method. In these cases the density on the CT is mot a true representation of the tissue density at the time of treatment. However, the physicist may determine that a correction is not necessary if the magnitude of the effect on the dose distribution is small. By similar triangle geometry, 40 at 100 = 160 at 400. The field size at 88 cm SSD on the skin, X, is given by similar triangle geometry: X = 15 (88/100) = 13.2 cm. The dose to the fetus depends on its distance from the field edge, but from 10 to 20 cm the dose at 10 cm depth is between about 2% and 0.6% of the dose on the beam axis. (Ref: AAPM Report No. 50, “Fetal Dose from Radiotherapy with Photon Beams,” AAPM Radiation Therapy Committee Task Group 36, Reprinted from Med Phys 22(1):63-82, 1995.) The dose is made up of patient scatter, head leakage, and radiation scattered from the collimators and blocking tray. MU = dose/(dose rate at midplane) x TMR = 150/[(1.0/4.5)? x OF x 0.87] = 3491 MU approx. (ignoring OF) Time = MU/MU per min = 3491/600 = 5.8 min approx. Although cones offer advantages in simplicity and reproducibility, the shapes they produce are essentially spherical, unless more than one isocenter is used. MLCs require constant maintenance and more complex QA, but offer the opportunity to shape fields conformally, thus reducing normal tissue dose. Raphex 2007 27Therapy + Answers T43. 744. T45. T46. T47. T48. T49. TS50. TSI. ICRU Report 62 defines the Internal Target Volume (ITV) as the volume formed by the Clinical Target Volume (CTV) and the Internal Margin (IM). The ITV represents the volume encompassed by the CTV as it moves with breathing and internal organ motion, and is distinct from the Setup Margin (SM). Both SM and IM are used to create the Planning Target Volume (PTV) from the CTV. Reducing the dose rate will reduce the number of beam holds. A “planning organ at risk volume” (PRV) is created by adding a margin around an “organ at risk” (OAR) to account for geometric uncertainty in its location. It is analogous to adding a margin around the CTV to create a PTV. The predominance of the Compton effect in megavoltage photon beams means that there is no differential attenuation on the basis of Z value, as in diagnostic beams, and it is only the difference in density of bone that allows it to be visualized. (Consequently, lungs and sinuses are more easily visualized.) Scatter decreases contrast but cannot be removed by a conventional grid since the penetrability of the beam would require very great thicknesses of lead and would not be feasible. Increasing mAs will make the film darker, but have no effect on contrast. Contrast is improved by reducing scatter. Reducing the collimator setting to the minimum necessary field of view usually has the greatest effect. If different grids are available, the one with the greatest grid ratio will “clean up” the most scatter. If these two techniques still do not work, for patients with very large lateral separations one solution is to take orthogonal R and L anterior obliques, making use of the somewhat reduced separation. Cerebrospinal fluid (CSF) appears dark on a T1 image, and bright on a T2 image. Findings of interest such as transependymal edema or intraventricular tumor can be obscured by the brightness of the CSF on a T2 image. FLAIR allows the CSF to appear dark, and thus make T2 changes more conspicuous. Deep inspiration breath hold is not reproducible unless the depth of inspiration is controlled, either by gating with patient feedback, or by spirometry. If a device of this type is not available, the maximum excursion of the target volume should be used as a basis for creating the PTV. 28 Raphex 2007Therapy ¢~ Answers @ T52. T53. T54. T55. T56. T57. T58. T59. T60. T6l. T62. T63. T64. Skin dose is higher for electrons than for photons, and increases as electron energy increases. Because of the photoelectric effect, absorbers with a higher Z (e.g., bone) have enhanced absorption compared to tissue in low-energy x-ray beams. Skin dose increases as energy increases. Output for electron beams depends on all the factors stated. MU = Dose/Output x PDD = 200/(1.02 x 0.9) = 218 MU. If there is a sudden discontinuity in the bolus, there may be hot and cold areas in the dose distribution in tissue below the discontinuity. 6-MeV electrons are stopped in 3 cm of water, or approximately 3 mm of lead. Hot and cold spots are inevitable when adjacent electron fields are abutted. Moving the junction as many times as practical during treatment will reduce the hot and cold spots accordingly. The screen is usually about 1 cm thick, and placed about 20 cm in front of the patient. It is used to increase large-angle scatter, and so improve dose uniformity. However, it also acts as an energy degrader and decreases depth dose. Although the gamma energies are different, the dose distributions in tissue around Ra and Cs sources are similar. A = Ao exp (0.693 x 7/17) = 0.75 Ao. Unlike curies or becquerels, the air kerma strength (AKS) gives an indication of the dose rate at a given distance from the source. Convert dose rate to cGy/min at 1m: Dose/min = 40,000 x 1/60 x 1/(10*) = 0.067cGy. Raphex 2007 29Therapy «> Answers T65. Because of the short distances involved in brachytherapy, dose rate falls off quickly with T66. 167. T68. T69. T70. T7I. 72. 773. 74, T75. T76. distance due to the inverse square law. For example, between 2 cm and 4 cm the dose rate is reduced to '/, of its original value. By comparison, the same 2-cm displacement in a linac photon beam (100 to 102 cm) would reduce the dose rate to 0.96 of its original value. The shorter half-life of Pd (17 days) requires a higher initial activity. The dose is deliv- ered in a shorter time; hence the total dose is reduced accordingly. The Total dose = Initial dose rate x T. = 1600 cGy. After two half-lives, the dose rate has dropped to '/, times the initial dose rate. The total dose delivered from this point on will therefore be '/, x 1600 cGy, and the dose delivered during the first two half-lives will be (1 - 14.) x 1600 = 1200 cGy. The dose rate at A is a function of activity/distance’. R ovoid to A = 5 cm, L ovoid to A = 3 cm, 1/57:1/3? = 9:25. For dose rates below 2 mR/hr, no action is required. After a minimum of 10 half-lives (20 months for !?51) the seeds must be surveyed with an appropriate instrument to ensure that the dose rate is not above background. They can then be discarded in regular trash, provided documentation is kept. Due to the inverse square law, large diameter ovoids decrease the mucosal surface dose but increase the depth dose. Since ovoids (except mini ovoids) have the same internal shield- ing, doses to bladder and rectum are not affected. The treatment time will increase as the source strength decays with time, but the product of source strength and time will remain constant. 30 Raphex 2007Therapy ¢ Answers 177. T78. T79. T80. T8iI. T82. T83. T84. T85. T86. T87. T88. The dose is prescribed at 1 cm from the balloon’s surface. It is not necessary to measure the source activity before each use. Decay tables are generated at the time of source exchange and calibration, and these are used to verify the activity displayed at the treatment console before treatment. Seeds are ordered for a specific source strength on the day of the implant, and all dosimetry calculations are based on this. Magnetrons generate microwaves and are usually used in the lower energy range. Klystrons are microwave amplifiers. They are more expensive than magnetrons but last longer and are generally used in the higher energy range. A monitor unit (MU) represents the amount of charge collected by the monitor ionization chamber (located in the path of the beam, in the head), when a specific dose has been delivered at a reference point in a phantom. The beam terminates after the set number of MUs have been delivered. As the dose rate on a linac is not absolutely constant, dose cannot be monitored by time alone. However, linacs have a backup timer in case the monitor (and backup) chambers fail to terminate the beam. Beam flatness and symmetry are checked by the physicist during the monthly QA. It is good practice to calibrate the output monthly and to adjust the calibration of the monitor chamber if it has drifted by more than 2%. Daily output checks should also be done by a fast, simple method. If they are outside a given range, an investigation is required to solve the problem. Raphex 2007 31Therapy “+ Answers T89. T90. The equation for the transmission factor “B” is: B = Pd?/WUT. The half-value layer (HVL) of the barrier material will be needed to calculate the barrier thickness, but it is not required to calculate B. T9I. The rule of thumb is that the dose scattered laterally, from a megavoltage beam, has a maximum photon energy of about 500 keV, and the dose at 1 m is about 1/1000th of the dose at the isocenter. T92. The annual dose limit to a radiation worker’s hands is 500 mSv (50 rem). [See NCRP Report No. 116, “Limitation of Exposure to Ionizing Radiation,” 1993. (Supersedes NCRP report No. 91).] T93. T94. The voltage applied to an ion chamber makes it unsafe for in-vivo measurement, in case of a short. T95. This would cause an overdose of about 30% (1/TMRipo x 10, a0) “A” would deliver the dose 50 cm short of the target—probably in air. SEE ee el 32 Raphex 2007