1 Medical Physics Radiation Risk Garringer and Aliison Dr. Muhammad Bin Zulfiqar

1

Medical Physics: Radiation Risks DR. Muhammad Bin Zulfiqar PGR-FCPS III SIMS/SHL

GRAINGER & ALLISON’S DIAGNOSTIC RADIOLOGY

• FIGURE 1-1 The generally accepted LNT model implies ■that all exposures, even if they are very low, carry a certain risk. It is sometimes suggested that there may be some protective effect of radiation at very low doses but other studies show hyperradiosensitivity at very low doses

• FIGURE 1-2 Dose distribution during ■fluoroscopy at the operator’s position 80 cm from a water phantom placed on the table for under- and overcouch X-ray tube configurations. Operating conditions: 125 kVp, 1 mA.

• FIGURE 1-3 Chart representing the relationship between dose and relative average diameter ■as measured by the scout view (topogram) used by AEC systems to adapt the dose to the patient’s weight. The strongest increase in dose is expected with AEC warranting a constant noise (GE and Toshiba CT systems), whereas the dose increase may be moderate with AEC tolerating more noise in obese patients (Siemens and Philips CT systems).

• FIGURE 1-4• ■ Brain CT performed in two consecutive acquisitions obtained at 120

kV (64 × 0.6 mm) in a 70-year-old man with acute stroke. Arrows show a hypoattenuated area in the left parietotemporal region. Image thickness is 3 mm. (A) was obtained at 120 kV and 300 mAseff with AEC switched off and the corresponding CTDIvol is 63 mGy. (B) was obtained at 120 kV and 280 mAseff (quality reference index) with AEC switched on. The tube current–time product is lowered at 198 mGy by the AEC system and the corresponding CTDIvol is 42 mGy. The reconstruction filtered back projection kernel is H20, slightly smoother than the filtered back projection H30 used in (B).

• FIGURE 1-5 ■ Two consecutive brain CT examinations in a 42-year-old man who had a motor vehicle accident. Acquisitions were obtained at 120 kV and with a collimation of 16 × 0.6 mm. Image thickness is 3 mm. (A) was obtained with fixed tube current at 450 mAseff and resulted in a CTDIvol of 61 mGy (reconstruction filtered back projection kernel is H30). (B) was obtained 3 months after (A) with AEC switched ‘on’, a quality reference mAs at 380 mAs and mean effective tube current–time product of 312 mAs, resulting in a CTDIvol reduced to 42 mGy (reconstruction filtered back projection kernel H20). An arachnoid cyst is seen in the left temporal region. ;

• FIGURE 1-6 ■ Optimised brain CT acquisition obtained with 40 × 0.6 mm collimation, 140 kV, AEC switched ‘on’, a quality reference mAs at 170. The resulting delivered tube current–time product is 114 mAseff and the CTDIvol is 30 mGy. DLP is 442 mGy.cm, approximately one-half of the RDL for brain CT that is usually set at 1000 mGy.cm. (A) Axial 2-mm slice and (B) coronal 3-mm slice reconstructed with filtered back projection kernel H20.

• FIGURE 1-7 ■ Optimised brain CT acquisition in an 8-year-old boy with 100 kV, activated AEC and quality reference tube current–time product at 300 mAseff. The resultant tube current– time product is 237 mAseff and the CTDIvol at 20.7 mGy. Slice thickness is 3 mm and reconstruction filtered back projection kernel is H20.

• FIGURE 1-8 ■ Comparison of two consecutive CT images of sinunasal cavities obtained from a 34-year-old woman, including axial and coronal 2-mm-thick CT slices. The first examination (A) was acquired in 2010 on a 16-slice MDCT system with a CTDIvol of 6.75 mGy and a DLP of 66 mGy.cm. The second examination (B) was acquired on a 2 × 64-slice MDCT system with a CTDIvol of 3.3 mGy and a DLP of 40 mGy.cm. Kernel is a filtered back projection algorithm for both examinations (H50).

• FIGURE 1-9 ■ Axial 3-mm slices at the level of lung bases obtained in two male patients weighing 85 kg, with similar chest wall thickness and 33 cm in lateral chest diameter. On the left, the acquisition parameters were not optimised and the noise index specific to the AEC system was set at 11 HU. The resulting CTDIvol is of 28.3 mGy and exceeds the French national DRL (set at 15 mGy for chest CT). On the right, acquisition parameters were optimised and the AEC system tolerated a noise of 15 HU and limited the dose increase from 80 to 100 mAseff. The resulting CTDIvol is 7.7 mGy.

• FIGURE 1-10 ■ Consecutive chest CT examinations in the same patient with stable body weight of 65 kg obtained in 2010 and in 2011. A pulmonary lesion appeared in 2011 in the left upper lobe. The 2010 acquisition was obtained with 120 kV and 80 mAs, whereas the second CT was obtained at 140 kV and 38 mAs default setting. The CTDIvol was reduced from 4.24 to 2.36 mGy (–45%). Image quality is very similar between both acquisitions, as shown in mediastinal window in (A), and in pulmonary windows in coronal (B), axial (C) and sagittal views (D). C D A B

• FIGURE 1-10 ■ Consecutive chest CT examinations in the same patient with stable body weight of 65 kg obtained in 2010 and in 2011. A pulmonary lesion appeared in 2011 in the left upper lobe. The 2010 acquisition was obtained with 120 kV and 80 mAs, whereas the second CT was obtained at 140 kV and 38 mAs default setting. The CTDIvol was reduced from 4.24 to 2.36 mGy (–45%). Image quality is very similar between both acquisitions, as shown in mediastinal window in (A), and in pulmonary windows in coronal (B), axial (C) and sagittal views (D). C D A B

• FIGURE 1-11 CTPA examination in a patient weighing 55 ■kg obtained at 80 kV and with a quality reference mAs AEC setting at 12 mAs, reduced by the AEC to 56 mAseff in the present patient. The CTDIvol of this acquisition is 1.1 mGy and the DLP of the entire examination is 39 mGy.cm. The use of a 80-kV tube potential reduces the dose and makes it possible to obtain a perfect vessel opacification.

• FIGURE 1-12 ■ CTPA examination in a patient weighing 115 kg obtained at 120 kV. The CTDIvol of this acquisition is 10.5 mGy and the DLP 394 mGy.cm. The vessel enhancement is not excellent, probably because of tube potential at 120 kV. A pulmonary embolism is seen in the right upper lobe. Note that this obese patient was exposed to a DLP that does not exceed the RDL of a standard patient.

• FIGURE 1-13 ■ Low-dose chest CT examination acquisition obtained in a 23-year-old woman with suspected tuberculosis. Tube potential is 140 kV, quality reference mAs (and the tube current is 9 mAs) of AEC is set at 9 mAs and the CTDIvol is 0.97 mGy, whereas the DLP is 33 mGy.cm. Slice thickness is 3 mm and reconstruction kernel uses iterative technique. As shown in Fig. 1-14, in a standard patient weighing 70 kg, such low-dose chest CT protocol delivers a DLP not higher than 20 mGy.cm. Using 0.017 mSv/mGy.cm as conversion factor, the effective dose is 0.34 mSv, equivalent to the dose of a two-view radiographic examination.

• FIGURE 1-14 ■ Low-dose chest CT examination obtained in a patient weighing 65 kg with the same tube potential and quality reference mAS as in Fig. 1-13. AEC system reduced the tube current from 9 to 5 mAseff. The CTDIvol is of 0.57 mGy and the DLP of 18 mGy.cm. Note the tracheobronchial diverticuli.

• FIGURE 1-15 ■ Optimised dose unenhanced MDCT obtained in an obese patient weigting 135 kg, with acute appendicitis. The high tube potential and low tube current strategy is used (140 kV–120 mAs). CTDIvol is 15 mGy and dose–length product is 747 mGy.cm. Average abdominal diameter is 40 cm.

• FIGURE 1-16 ■ Optimised dose 16-detector row CT (Emotions 16, Siemens, Forchheim, Germany) in a 68-year-old man weighing 75 kg (standard size patient) and with right lower quadrant pain. Image noise measured in inferior vena cava is 16 HU. Tube potential is 130 kV, quality reference mAs at 80 mAseff and is lowered to 50 mAseff by the AEC. The subsequent CTDIvol is 5.6 mGy. Acquisition height is 37 cm and the corresponding DLP is 233 mGy.cm.

• FIGURE 117 ■ Reduction of tube potential in a patient weighing 67 kg with chronic pancreatitis and a pseudocyst in the head of the pancreas. On the left, 120-kV MDCT with default 120-mAs quality reference tube current. The CTDIvol is 4.2 mGy and the DLP is 194 mGy.cm. On the right, tube potential was set at 100 kV with unchanged quality reference tube current (120 mAs). The CTDIvol is reduced to 2.5 mGy and the DLP is 101 mGy.cm. For the examinations, the AEC system has adjusted the tube current from 120 to 62 mAseff.

• FIGURE 1-18 ■ Low-dose unenhanced MDCT of the abdomen in an extremely obese 22-year-old woman with right iliac fossa pain. Tube potential is 140 kV and the quality reference mAs at 30 mAseff. AEC system adapted the tube current–time product to 57 mAs. The CDTIvol is 5.9 mGy, whereas the DLP is 290 mGy.cm. Arrow shows the normal appendix.

• FIGURE 1-19 ■ Low-dose unenhanced MDCT of the abdomen in an obese 21-year-old, 1.78-m-tall man, weighing 105 kg referred for acute pain in the right iliac fossa. Tube potential is 140 kV and quality reference tube current is set at 20 mAseff. AEC increased this default tube current to 35 mASeff. The CDTIvol is 3.7 mGy, whereas the DLP is 131 mGy.cm. Note that this low DLP was achieved because the acquisition height has been limited to the lower three-quarters of the abdomen. A standard acquisition would have delivered 900 mGy.cm in this patient.

• FIGURE 1-20 ■ Low-dose unenhanced MDCT of the abdomen in a 21-year-old man weighing 90 kg and 1.78 m tall, with suspected acute left renal colic. Tube potential is 140 kV and quality reference tube current is set at 30 mAseff. AEC adapted the tubecurrent–time product to 32 mAseff. The CDTIvol is 3.4 mGy and the DLP is127 mGy.cm. A 2-mm large calculus is seen in thedistal left ureter.

• FIGURE 1-21 ■ Low-dose iodine-enhanced MDCT of the upperabdomen with a 21-cm acquisition height performed in order to confirm acute right pyelonephritis (arrow) in a 22-year-old woman. Tube potential is lowered at 110 kV and the quality reference tube current is set at 60 mAs, reduced to 28 mAs b the AEC. The applied CTDIvol is 2 mGy and the DLP is limited to 42 mGy.cm, equivalent to the risk of an abdominal plain radiograph.

• FIGURE 1-22 ■ Iodine-enhanced abdominal MDCT in a 12-year-old boy weighing 45 kg complaining of right iliac fossa pain. No sign of appendicitis was found by CT. Tube potential is 100 kV and quality reference tube current is 120 mAseff. The AEC reduced this tube current to 57 mAseff. The resulting CTDIvol is 2.6 mGy and the DLP is 101 mGy.cm. (A) A representative 3-mm axial view and (B) coronal and sagittal 3-mm views.

• FIGURE 1-23 ■ Low-dose unenhanced MDCT of the abdomen obtained in a 11-year-old boy weighing 40 kg, referred for right iliac fossa pain after inconclusive US examination. Tube potential is 100 kV and quality reference tube current at 70 mAseff. This preset is reduced to 37 mAseff by the AEC system. The resulting CTDIvol is 1.7 mGy and the DLP is 51 mGy.cm. The appendix (arrows) is normal in axial (A) and in coronal and sagittal orientations (B).

• FIGURE 1-24 Unenhanced MDCT obtained at ■80 kV in a 9-yearold boy weighing 34 kg after inconclusive ultrasound examination showing an acute appendicitis (arrow) while delivering 1.4 mGy (CTDIvol) and 50 mGy.cm (DLP).

• FIGURE 1-25 ■ Comparison of two consecutive CT-enhanced acquisitions of the abdomen with a dose reduction of 33% in an 87-yearold woman weightng 88 kg. The first one is performed with 120 kV and 150 mAseff, and the second one with the same tube potential but 100 mAs. Slice thickness is 3 mm. Image noise is slightly higher in the 100-mAs axial (A) and coronal (B) orientations, but the 100-mAs images are of acceptable quality.

• FIGURE 1-26 ■ Comparison of two consecutive unenhanced MDCT examinations in a 48-year-old man weighing 75 kg who had a sigmoid perforation on a foreign body that proved to be a swallowed toothpick. The toothpick is visible on the right-sided coronal reformat. The first CT was not optimised and at 120 kV and a noise index of 10 UH for 1.25-mm slices. The resultant CTDIvol is approximately 20 mGy and the DLP is 1009 mGy.cm. The second CT displayed on the left was acquired after endoscopic removal of the toothpick at 140 kV and a quality reference tube current of 40 mAseff, reduced to 36 mAseff by AEC. The corresponding CTDIvol was 3.8 mGy and the DLP was 148 mGy.cm, the dose being reduced by a factor of 7.

• FIGURE 1-27 ■ Side-by-side comparison of two consecutive unenhanced MDCT of the lumbar spine obtained in a 67-year-old patient weighing 92 kg, with stage IV colon carcinoma and complaining of low-back pain. Two acquisitions are obtained, one at standard dose of 68 mGy (CTDIvol, displayed on the right), and the second at an optimised dose of 36 mGy (CTDIvol, displayed on the left). (A) Sagittal reformats in soft-tissue algorithm and window, (B) axial slices and (C) sagittal reformats with bone algorithm and window.

• FIGURE 1-27 ■ Side-by-side comparison of two consecutive unenhanced MDCT of the lumbar spine obtained in a 67-year-old patient weighing 92 kg, with stage IV colon carcinoma and complaining of low-back pain. Two acquisitions are obtained, one at standard dose of 68 mGy (CTDIvol, displayed on the right), and the second at an optimised dose of 36 mGy (CTDIvol, displayed on the left). (A) Sagittal reformats in soft-tissue algorithm and window, (B) axial slices and (C) sagittal reformats with bone algorithm and window.

• FIGURE 1-28 ■ Side-by-side comparison of two consecutive unenhanced MDCT of the lumbar spine obtained in a 72-year-old man weighing 71 kg, with stage III non-small cell lung carcinoma who complained of low-back pain. Two acquisitions are obtained, one at a standard dose of 38 mGy (CTDIvol, displayed on the right), and the second at an optimised dose of 23 mGy (CTDIvol, displayed on the left). (A) Axial slices at the level of L5–S1 disc showing a disc herniation. (B) The same herniation (arrow) in left parasagittal orientation. Sagittal reformats in soft-tissue algorithm and window.

• FIGURE 129 ■ Side-by-side comparison of sagittal reformats of the lumbar spine showing the potential benefit of iterative reconstruction technique for imaging low-back pain with CT. The right one is obtained with iterative reconstruction and the left one with usual filtered back projection technique. Noise seen in FBP reformat is significantly reduced by the iterative technique.