IN RECENT years, the commercialization of the multislice computed tomography (MSCT) scanner and the 256‐MSCT, in particular, has shortened CT scanning and imaging time, reduced motion artifacts when examinees breathe, and enhanced image resolution. MSCT is currently the most commonly used equipment in diagnostic radiographic examinations. But the radiation dose given to MSCT patients is also relatively higher than the dose from other diagnostic examination. According to National Council on Radiation Protection and Measurements (NCRP) Report 160 (2009), statistical data from 2006 show that the medical exposure of Americans accounted for nearly 50% of the total population’s radiation exposure, of which the medical exposure from CT examinations alone accounted for as much as 24% of the total population’s radiation exposure. Therefore, the problem of reducing radiation dose from CT examinations has become a public concern since CT is increasingly and widely being adopted as a diagnostic tool.
Effective doses from MSCT scans vary with the type of CT examination, which is also related to detector rows, effective tube current-time product, tube voltage, pitch, and many other factors. During the CT examination, organ doses are often assessed by placing thermoluminescent dosimeters (TLDs) in locations of relevant organs in an anthropomorphic phantom, such as the Rando phantom (Phantom Laboratory, Cambridge, NY). It is a convenient and effective measurement method with the capability of measuring multipoint doses simultaneously. Doses to various organs based on measurement are then multiplied by the tissue-weighting factor (wT) provided by International Commission on Radiological Protection (ICRP) Report 103 (2007) to assess effective doses given to patients during the CT examination (Chang et al. 2010). However, there has been less research on abdominal examinations. The effective dose from these examinations is important for the patient with advanced cancer who must receive radiotherapy. Effective doses are convenient comparison indicators used to compare patients’ doses from CT phantom examination or actual CT examinations.
There have been many research articles published on patients’ radiation doses from CT scans measured with TLDs (Cheung et al. 2007 ; González et al. 2007 ; Tsai et al. 2007 ; Tyan et al. 2008). Nishizawa et al. (2007) compared three types of 4‐detector row CT scanners and three types of 16‐detector row CT scanners to examine doses to adults’ lungs and abdomens and children’s heads and to assess their effective doses with routine clinical checks. Mori et al. (2006, 2008) compared patients’ organ doses and effective doses caused by cardiac imaging using c-arm CT (CACT) with 16‐MSCT, 64‐MSCT, and 256‐MSCT scans. It was concluded that 256‐MSCT scans can provide less effective doses than 16‐MSCT and 64‐MSCT scans. Moreover, it can further reduce the dose when used together with automatic exposure control systems (AECs).
In terms of commonly used machines, the 256‐MSCT scanner is a major piece of CT equipment increasingly and widely used in clinical diagnosis and medical examinations. While there are many papers that mainly discuss the effective doses caused by CT examinations of the head and neck, as well as low-dose chest examinations and cardiovascular scans (Walker et al. 2009), there has been less research on abdominal examinations. This research attempts to compare effective doses from simulated spine tumor examinations using 64‐MSCT and 256‐MSCT scans. In addition to adopting a routine clinical protocol check, we deliberately changed the conditions of pitch, tube voltage (kV), and effective tube current-time product (mA s) when measuring organ doses. We then converted them to effective doses and compared the difference in order to provide practical references for clinical diagnosis. As we researched the change in radiation doses by changing the operating parameters, we also analyzed the image spatial resolution, and we discuss the impact on the image quality and diagnostic information.
MATERIALS AND METHODS
CT and scan conditions
The study used two types of CT scanners of the same brand—namely, a 64‐MSCT (Brilliance 64; Philips, Andover, MA, U.S.) scanner and a 256‐MSCT (Brilliance iCT; Philips) scanner—to scan the spine with a routine protocol, as summarized in Table 1. The scanners were used to measure absorbed doses (mGy) to each organ by halving the pitch, lowering the effective tube current-time product (from 250 to 150 mA s), and raising the tube voltage (from 120 to 140 kV), and to assess effective doses using the tissue-weighting factors (wT) of ICRP 60 (1991). The scanning protocols and the exposure parameters of these two kinds of CT scanners are shown in Table 1, with the experimental conditions of the 256‐MSCT scanner labeled as item (condition) A, B, C, D, and of the 64‐MSCT scanner labeled as item (condition) E, F, G, H.
The absorbed doses to organs and skin were measured by taking data from an anthropomorphic Rando phantom. This study placed TLDs in the relevant positions of the Rando phantom’s surface and various internal organs. In addition to using the organs of ICRP 60 (1991) as references when placing TLDs inside the phantom, we also measured doses to four other organs—brain, salivary gland, small intestines, and rectum—and classified these as remainder organs. The scan covered the whole abdomen from the diaphragm to the pubic symphysis with a scan length of 450 mm.
Meanwhile, the CT scan conditions changed in pitch, tube voltage (kV), and effective tube current-time product (mA s). Because image spatial resolution has become a critical factor in medical diagnosis, this study also used an American College of Radiology (ACR) CT accreditation phantom (model number 464; Gammex Inc., Middleton, WI, U.S.) to experiment with image spatial resolution under various scan conditions (conditions A to H) listed in Table 1. In terms of diagnostic requirements, the spatial resolution of each experimental condition can be accepted only when it is more than 5 line pairs per centimeter (lp cm−1); that is, when the spatial resolution capacity can resolve the line that is closer to the spacing distance of 1 cm (Mahesh 2009).
This study adopted two kinds of TLDs; namely, GR‐200 and GR‐200F (LiF: Mg, Cu, P), both of which have relatively high sensitivity to photon radiation. The GR‐200 TLD is a 3.2 × 3.2 × 0.89 mm chip and can be placed inside the Rando phantom. The film TLD GR‐200F is a detector 5 mm in diameter and 0.1 mm in thickness and can be affixed to the surface of the phantom. A simulated MSCT scan was performed on examinees with different detector rows, pitch, tube voltage (kV), and effective tube current-time product (mA s) parameters so as to measure the internal organ doses and skin doses (Lin et al. 2001). The TLD annealing temperature remained at 240°C for 10 min, and the Harshaw 3500 TLD reader (ThermoFisher Scientific, Waltham, MA, U.S.) was set for preheating at 50°C, a heating rate of 10°C s−1, and 240°C when reading the TLDs.
Dose-response linearity and energy dependence of each TLD were calibrated with a 137Cs standard source from the National Radiation Standard Laboratory, Institute of Nuclear Energy Research, Taiwan. The average error of the dose-response factor of each TLD was selected to obtain results that were less than ±3% for GR‐200F TLDs and less than ±1.5% for GR‐200 TLDs. Before the CT scan measurement, the labeled TLD was placed in a particular organ location in the phantom and the placement was recorded to avoid confusion.
Since tube voltage in routine CT scan examinations is set at 120 kV for a general body size, the study used different voltages (kV) and tube current-time products (mA s) to obtain the best image spatial resolution for different body sizes. The TLD energy dependence was calibrated with an x ray (ISO 1996) of M120 beam quality relative to the standard radiation source of 137Cs (with 662 keV gamma radiation) to obtain the correct dose assessment value. The M120 x-ray nominal voltage was 120.7 kV, and the effective energy was 48.9 keV. With this energy, the dose response of the TLD was higher than the standard energy of 137Cs by 3–7%, the correction factor for average energy dependence was 0.94 ± 0.03, and thus the reading of the TLD dose was multiplied by 0.94 to get the correct absorbed dose in air (Poudel et al. 2015).
To assess dose to the organs in ICRP Report 60 having tissue-weighting factors, we placed a total of 39 GR‐200 TLD chips in their relative locations inside the Rando phantom, as shown in Fig. 1. TLD measurements of organ doses depend on the size and volume of the organs. Bone marrow and bone surface points had seven TLD locations not marked in Fig. 1.
The skin dose was measured using the film TLD, which was placed on the phantom surface and covered with 1.14 mg cm−2 polyethylene (PE) film. A total of 18 GR‐200F films were placed every 5 cm along the iliac crest in the transverse direction and from the eyebrows to the pubic symphysis in the longitudinal direction. The above organ doses and skin doses were all measured with three repeated experiments.
RESULTS AND DISCUSSION
Each TLD in the experiment had a calibrated dose-response factor, which was carefully recorded during repeated measurements, and the average error of each organ dose obtained was between ±0.5% and ±11%. Through this complex experimental process, we were able to obtain better measurement precision. In addition, as the dose-response factor of each TLD was calibrated with a 137Cs standard radiation source in the National Radiation Laboratory and was amended by the beam quality of M120 x rays from the ISO 4037‐1 report (1996), we are highly confident in the accuracy of the measurements made in this experiment.
A Rando adult phantom was used in the simulated spine CT examination. The operating parameters of the iCT 256‐MSCT scanner and the 64‐MSCT scanner are shown in Table 1, and the measured organ absorbed doses (mGy) and effective doses (mSv) obtained are shown in Table 2. The absorbed doses to various organs or tissues caused by the experimental conditions A–E are shown in Fig. 2 according to the top-down location in the body. The contributions of each organ to the effective doses are shown in Fig. 3. The trend in results from conditions F–H is similar to Fig. 2 and Fig. 3. Table 2 shows that effective doses from 256‐MSCT scans were significantly greater than those from 64‐MSCT scans, and the effective doses from routine operating conditions A and E were 25.2 mSv and 19.56 mSv, respectively. Although the scan time was shortened by 2.6 times, the effective doses from 256‐MSCT scans were still higher than those from 64‐MSCT scans by as much as 22%. Differences between organ doses within the scanned area, which includes gonads, colon, stomach, bladder, liver, and the skin, were less than 25%. But differences between organ doses outside the scanned area, which includes bone marrow, lung, breast, esophagus, and thyroid gland, were more than 27%. This was because the greater the quantity of detector rows, the larger the contribution of the penumbra dose. In terms of different pitches, effective doses from condition B and condition F decreased by 9% [(A − B)/A] and 3% [(E − F)/E], respectively. Thus, it is apparent that halving the pitch not only leads to an increase in scan time, but also results in a slight decrease in the majority of organ doses and effective doses when effective tube current-time product (mA s) and tube voltage (kV) remain unchanged. After examination, it was found that the CT dose index (CTDI) values for pitch 0.507 (condition B) and 0.516 (condition F) were indeed lower than those for conditions A and E, which agreed with the trend in measured values from this experiment.
When the effective tube current-time product was reduced from 250 mA s (condition A) to 150 mA s (conditions C and G), effective doses remained at 14.38 mSv and 11.99 mSv, respectively, and substantially declined by 43% [(A − C)/A] and 39% [(E − G)/E], respectively. The trend of the two types of scanners was the same, which indicated that lowering the effective tube current-time product (mA s) decreases the number of x-ray photons and decreases organ doses, including a significant decline in the skin dose. The tube voltage (kV) value rose from 120 kV (condition A and condition E) to 140 kV (condition D and condition H), but the mere increase of 20 kV (17%) could enhance the penetration power due to the increase in the average energy of photons, with effective doses of 36.48 mSv and 29.78 mSv as well as an increase of 45% (condition D) and 52% (condition H). The results show that changing the effective tube current-time product (mA s) and tube voltage (kV) would significantly change the radiation dose to the phantom, while changing the pitch has a smaller impact. It is seen in Fig. 2 and Fig. 3 that among the five scanning conditions from condition A to condition E in the simulated spine tumor scan examination, gonads greatly contribute to effective doses, followed by the colon and stomach; the total contributions to the effective dose (mSv) of these three organs from condition A reached as much as 61%. Similarly, those from condition E accounted for as much as 63%. In addition, among the four remainder organs, the organ doses of small intestines and the rectum were 35.51 and 26.44 mGy, respectively, from condition A and 28.40 and 21.00 mGy, respectively, from condition E, which were second only to the gonad doses of 36.10 mGy (condition A) and 30.07 mGy (condition E). Nevertheless, since these two were averaged with the lower brain and salivary gland doses and then multiplied by wT = 0.05, their contribution to the effective dose was only about 3%.
Mettler et al. (2008) determined that effective doses that resulted from abdominal CT examinations ranged from 3.5 to 25 mSv, with pelvis doses at 3.3–10 mSv. They also suggested using appropriate conditions to obtain better images and reduce patient doses with different scanning positions. Effective doses to the whole abdomen measured in this study ranged between 14.38 and 25.2 mSv, which were consistent with other previous study results. In a similar study, Tyan et al. (2008) employed 64‐MSCT scans to check patients’ whole abdomens and assess the actual effective doses. The result showed that doses to males ranged from 20.6–34.6 mSv and females from 21.3–55.2 mSv, with the average being significantly higher than the average in this study.
Changes in CT operating conditions to reduce effective doses must be based on good image quality which is a diagnostic requirement, otherwise they are meaningless. With the same brand of CT scanner but different models, this study used the ACR CT accreditation phantom (model number 464) to obtain a high-contrast, high spatial-resolution image with eight abdomen protocols shown in Table 1. All of these conditions can attain a clear resolved capacity of 5 lp cm−1 to 6 lp cm−1. In Table 2, lowering the effective tube current-time product from 250 to 150 mA s (as in condition C and condition G) can decrease the effective dose by 39–43%, but raising the tube voltage from 120 to 140 kV (as in condition D and condition H) can increase the effective dose by 44–52% while there is no significant change in spatial resolution. Therefore, when performing a CT scan, the operating conditions of scanners can be slightly modified (such as reducing tube current-time product or tube voltage) based on the patient’s body without affecting the image resolution. This would likely reduce the radiation dose to examinees and would be a great help in reducing the population’s medical radiation doses.
This study used an anthropomorphic phantom with two types of MSCT scanners of the same brand; namely, 256‐MSCT and 64‐MSCT scanners. It attempted to identify differences in examinees’ effective doses from a simulated spine tumor examination scan under different operating conditions such as routine clinical operating protocols and changes in pitch, effective tube current-time product (mA s), and tube voltage (kV). The results show that effective doses from 256‐MSCT scans are higher than those from 64‐MSCT scans. Changing both the effective tube current-time product (mA s) and the tube voltage (kV) are two key factors that affect doses. Despite changes in operating conditions for MSCT scans, the spatial resolutions achieved the capability of 5 lp cm−1 to meet diagnostic needs. Therefore, physicians and medical physicists should adjust for examinees’ physical (body weight) conditions and scanning areas to reduce patient doses, without affecting diagnostic requirements, by selecting the appropriate pitch, effective tube current-time product (mA s), and tube voltage (kV).
This study was supported by Tri-Service General Hospital, National Defense Medical Center, Taiwan (TSGH-C105‐080).
Chang Y-L, Lee C-M, Hsiao W-T, Hsu F-Y, Kuo M-J, Chiu J-H. Dose
evaluation of multi-slice CT for different parameters in chest examinations using TLD method. Radiat Meas 45:701–703; 2010.
Cheung T, Cheng Q, Feng D. A practical method for determining organ dose
during CT examination. Appl Radiat Isot 65:239–242; 2007.
González P, Furetta C, Azorin J. Comparison of the TL responses of two different preparations of LiF: Mg, Cu, P irradiated by photons of various energies. Appl Radiat Isot 65:341–344; 2007.
International Commission on Radiological Protection. 1990 recommendations of the International Commission on Radiological Protection. Oxford: ICRP; Publication 60; 1991.
International Commission on Radiological Protection. The 2007 recommendations of the International Commission on Radiological Protection. Oxford: ICRP; Publication 103; 2007.
International Standards Organization. X and gamma reference radiation for calibrating dosemeters and dose
rate meters and for determining their response as a function of photon energy—Part 1: Radiation characteristics and production methods. Geneva: ISO; ISO 4037‐1; 1996.
Lin J-P, Chu T-C, Lin S-Y, Liu M-T. Skin dose
measurement by using ultra-thin TLDs. Appl Radiat Isot 55:383–391; 2001.
Mahesh M. MDCT physics: The basics—technology, image quality and radiation dose
. Baltimore, MD: Lippincott Williams & Wilkins; 2009.
Mettler F, Huda W, Yoshizumi T, Mahesh M. A catalog of effective doses in radiology and nuclear medicine. Radiol 248:254–263; 2008.
Mori S, Nishizawa K, Kondo C, Ohno M, Akahane K, Endo M. Effective doses in subjects undergoing computed tomography
cardiac imaging with the 256‐multislice CT scanner. European J Radiol 65:442–448; 2008.
Mori S, Nishizawa K, Ohno M, Endo M. Conversion factor for CT dosimetry
to assess patient dose
using a 256‐slice CT scanner. Br J Radiol 79:888–892; 2006.
National Council on Radiation Protection and Measurements. Ioninizing radiation exposure of the population of the United States. Bethesda, MD: NCRP; NCRP Report 160; 2009.
Nishizawa K, Mori S-I, Ohno M, Yanagawa N, Yoshida T, Akahane K, Iwai K, Wada S-I. Patient dose
estimation for multi-detector-row CT examinations. Radiat Protect Dosim 128:98–105; 2007.
Poudel S, Currier B, Medich DC. Variation in the calibrated response of LiF, Al2
, and silicon dosimeters when used for in-phantom measurements of source photons with energies between 30 keV and 300 keV. Health Phys 108:434–442; 2015.
Tsai H, Tung C, Yu C, Tyan Y. Survey of computed tomography
scanners in Taiwan: Dose
guidance levels, and effective doses. Med Phys 34:1234–1243; 2007.
Tyan Y, Tsai H, Hung Y, Lia N, Chen C. In vivo dose
assessment of multislice CT in abdominal examinations. Radiat Meas 43:1012–1016; 2008.
Walker MJ, Olszewski ME, Desai MY, Halliburton SS, Flamm SD. New radiation dose
saving technologies for 256‐slice cardiac computed tomography
angiography. Int J Cardiovascular Imaging 25:189–199; 2009.