A survey of diagnostic reference levels for head,chest,and abdomen and pelvis CT in private diagnostic facilities in Norway

Abstract

Background

Computed tomography (CT) scans account for 60% of the total radiation dose in medical imaging. Literature has shown that patient dose varies across CT scanners, diagnostic protocols, and technical parameters at each site, suggesting an opportunity for starting an optimization process through establishing diagnostic reference levels (DRLs).

Purpose

To establish local DRLs (LDRLs) for six Norwegian private diagnostic institutes for frequently performed CT protocols.

Material and Methods

Dose data from 900 patients were collected from six diagnostic facilities. Data were recorded from non-contrast CT scans of the head and contrast-enhanced scans of the thorax and abdomen and pelvis from average-sized adult patients. An ANOVA test was performed to determine the variation in dose between scanners. LDRLs were determined by the 75th percentile of median values from dose indicators of CT scanners.

Results

The difference between the means of the dose distribution from each scanner was statistically significant (P < 0,05) for all examinations. The LDRLs determined were lower, for both national and international DRLs.

Conclusion

Observed dose variations from the scanners indicate a need for protocol optimization for some institutes, while the LDRLs demonstrate a potential for establishing newer national diagnostic reference levels (NDRLs) in Norway.

Keywords

Computed tomography dosimetry diagnostic reference levels dose optimization radiation protection

Introduction

In a bid to reduce the stochastic effects of ionizing radiation, there is a need to optimize the dose received from adult chest, abdomen/pelvis, and head computed tomography (CT) examinations to as low as reasonably achievable (ALARA) to obtain a lower dose to the patients with acceptable image quality (1). This protects patients from the harmful effects of radiation, thus ensuring radiation protection (1). The stochastic effects of ionizing radiation include heritable effects and malignant diseases (2). According to the American College of Radiology, the radiation dose to adults from a single CT scan is in the range of 1–15 mSv (3). CT scans contribute to 60% of the collective dose in medical imaging, which is disproportionate to the number of scans (3). Studies have shown that patient dose varies across regions and countries based on several factors, including scanner design, diagnostic protocols, and site-based choices of technical parameters (4). This implies there is a potential for optimization to reduce unnecessary radiation to patients (4). The most effective approach to reduce radiation exposure is to adapt the radiation dose to the patient´s body shape and size (5 –7). Depending on the vendor and CT scanner, this dose adoption can be performed through the use of automatic exposure control (AEC), such as SureExposure 3D (a standard deviation-based AEC), AutomA and SmartmA, and CAREdose4D (based on a set quality reference mAs) (8).

A proposal to investigate diagnostic reference levels (DRLs) for optimization of patient dose was initiated by the International Commission on Radiological Protection (ICRP) in 1996 (2). DRLs are tools used in medical imaging as a part of the optimization process for clinical exams (2). Collection of data from scanners indicates the dose received by an average-sized patient for the population being studied under routine conditions (4). DRLs allow hospitals to compare patient dose for CT procedures and if median doses exceed these DRLs, it can help to identify how that site can benefit from optimization (9). It is important to note that DRLs are only supposed to be used for optimization of protection for groups of patients and not individual patients (2).

By comparing doses used in routine CT protocols for average-sized patients, hospitals can use these data as a baseline tool in a continual Quality Assurance program, as DRLs are linked to departments or facilities (2). Establishing DRLs for specific protocols is one of the first steps in standardization and optimization to ensure that images maintain high diagnostic quality with a reduction in patient dose (3). DRLs set an upper threshold of dose, therefore facilitating identification of abnormally high doses within specific institutions (2). Implementation of DRLs has been shown to reduce overall dose to patients in the clinical setting (10).

Many countries with national DRLs (NDRLs) show reduced patient doses over time (11). Local DRLs (LDRLs) have also been established by groups of institutions or regions and can be compared to NDRLs (7). Using NDRLs or LDRLs as a guide, diagnostic departments can adjust CT scanning protocols according to patient characteristics to achieve a dose that is ALARA while still maintaining acceptable image quality (2,3).

The aim of the present study was to establish LDRLs for six Norwegian private diagnostic institutes for frequently performed CT protocols, including routine non-enhanced head scans, contrast-enhanced CT of the thorax and contrast-enhanced CT of the abdomen and pelvis. The registered CT doses for each body part and scan type will be compared with local, national, and international DRLs to identify the need for protocol optimization in the involved facilities.

Material and Methods

CT scanners and patient population

Dose data from the year 2022 were retrospectively recorded for a total of 900 patients using a TeamPlayer (Siemens Healthineers, Erlangen, Germany) based audit of data/CT archive. Data were recorded for three routine CT examinations from six private diagnostic facilities in Norway. A total of 50 adult patients were selected for each examination. CT dose index volume (CTDI_vol) and dose length product (DLP) data for average-sized patients scanned for each included CT examination with a standard CT protocol in the involved facilities were recorded. Identical CT protocols for each examination were included to make like-for-like comparison possible.

The study included routine non-contrast CT scans of the head, contrast-enhanced scans of the thorax, and contrast-enhanced scans of the abdomen and pelvis. Data were collected from four different scanner types for each of the abovementioned examinations: GE Discovery 690 PET/CT; Siemens Somatom Definition AS+; Siemens Somatom GoTop; and Toshiba Aquillion PRIME. Scans with standard protocols were included, and an overview of the characteristics of the scanners included in the study is shown in Table 1. Cases with a significant number of streaking artifacts present within the exam due to dental work such as fillings, crowns, or implants, were excluded. In addition, significantly mispositioned head exams, such as heads turned laterally, were also excluded. To ensure like-for-like comparison of the protocols, examinations using bolus tracking were also excluded.

Table 1.

Overview of characteristics of the CT scanners included in the study.

Institute	Manufacturer	Model	Install year	Detector configuration (mm)	Iterative reconstruction
1	GE	Discovery PET/CT 690	2010	64 × 0.625	ASIR
2	Toshiba	Aquillion PRIME	2016	80 × 0.5	AIDR3D
3	Siemens	SOMATOM Definition AS+	2014	64 ×0.6	SAFIRE
4	Siemens	SOMATOM Definition AS+	2015	64 × 0.6	SAFIRE
5	Siemens	SOMATOM Definition AS+	2014*	128 × 0.6	SAFIRE
6	Siemens	SOMATOM Definition AS+	2016	128 × 0.6	SAFIRE

*Differentiates between identical scanners from the same install year, located at different institutes.

Data collected include CT scan parameters, water equivalent diameter (WED/D_w), size-specific dose estimate (SSDE), CTDI_vol, and DLP. Size-specific dose estimate (SSDE) was developed to estimate the dose to the patient, and although the SSDE considers the geometric size of the patient, it does not account for the difference in X-ray attenuation when determining the radiation dose absorbed by the patient (12). Therefore, the American Association of Physicists in Medicine (AAPM) task group 220 developed a metric that can be used for estimating patient size, while also accounting for the difference in attenuation in the patient. This metric is known as the WED/D_w (12). According to the literature, the collected doses for DRLs should be based on average-size patients with a mean weight of 70 ± 10 kg (2,13). For adult patients, with unknown weight, the number of patients should be more than 20 for each particular examination (2). However, several studies show that values for WED representing an average-size patient are remarkably consistent, and it is thought that WED is a good surrogate for patient size. Hence, it can be used when collecting data for DRLs to select average-size patients (13 –15). For head protocols, all patients were considered of average size due to minimal size variation. For thorax and abdomen and pelvic examinations, cases with WED measuring 24.5–30 or 26.1–30 cm were selected as average-size patients, respectively (12,13). Information about equipment performance, scanner settings, and examination protocols are found in Tables 1, 2, and 3.

Table 2.

CT scan parameters of all examinations on each scanner.

Examination	CT Scanner	Tube voltage	AEC settings	Slice (mm)	Pitch	Rotation time (s)	Level of IR
Head
	Discovery PET/CT 690²⁰¹⁰	120 kV	Range = 140–335 mA, NI = 8	0625	0.531	0.8	–
	Aquillion PRIME²⁰¹⁶	120 kV	Range = 10–180 mA. SD = 1.5	0.5	0.625	0.75	–
	SOMATOM Definition AS+²⁰¹⁴	120 kV	200 ref mAs	0.6	0.55	1	SAFIRE3
	SOMATOM Definition AS+²⁰¹⁵	120 kV	220 mAs	0.6	0.8	1	SAFIRE2
	SOMATOM Definition AS+²⁰¹⁴*	120 ref kV, Care kV	380 ref mAs	0.6	0.6	1	SAFIRE3
	SOMATOM Definition AS+²⁰¹⁶	120 ref kV, Care kV semi	320 ref mAs	0.6	0.55	1	SAFIRE2
Thorax
	Discovery PET/CT 690²⁰¹⁰	120 kV	Range = 60–650 mA, NI = 38.3	0625	1.375	0.5	–
	Aquillion PRIME²⁰¹⁶	120 kV	Range = 120–500 mA. SD = 9	0.5	0.813	0.35	–
	SOMATOM Definition AS+²⁰¹⁴	120 kV	160 ref mAs	0.6	1.2	0.5	SAFIRE2
	SOMATOM Definition AS+²⁰¹⁵	120 ref kV, Care kV	70 ref mAs	0.6	1.2	0.5	SAFIRE3
	SOMATOM Definition AS+^2014*	120 ref kV, Care kV	110 ref mAs	0.6	1.2	0.5	SAFIRE3
	SOMATOM Definition AS+²⁰¹⁶	120 ref kV, Care kV semi	75 ref mAs	0.6	0.8	0.5	SAFIRE2
Abdomen and pelvis
	Discovery PET/CT 690²⁰¹⁰	120 kV	Range = 250–800 mA, NI = 34	0625	0.984	0.6	–
	Aquillion PRIME²⁰¹⁶	120 kV	Range = 110/500 mA, SD = 7.5	0.5	0.813	0.5	–
	SOMATOM Definition AS+²⁰¹⁴	120 ref kV, Care kV	200 ref mAs	0.6	0.6	0.5	SAFIRE3
	SOMATOM Definition AS+²⁰¹⁵	120 ref kV, Care kV	220 ref mAs	0.6	1	0.5	SAFIRE3
	SOMATOM Definition AS+²⁰¹⁴*	120 ref kV, Care kV	200 ref mAs	0.6	0.9	0.5	SAFIRE3
	SOMATOM Definition AS+²⁰¹⁶	120 ref kV, Care kV semi	160 ref mAs	0.6	0.8	0.5	SAFIRE2

Dash (–) indicates not applicable.

*Differentiates between identical scanners from the same install year, located at different institutes.

AEC, automatic exposure control; IR, iterative reconstruction; NI, noise index; SD, standard deviation.

Table 3.

Results of statistical analysis for the WED, SSDE, CTDI_vol, and DLP values of all examinations on each scanner.

CT scanner	Examination	WED (cm)	SSDE (mGy)	DLP (mGycm)	DLP (mGycm)	CTDI_vol (mGy)	CTDI_vol (mGy)
Discovery PET/CT 690²⁰¹⁰	Head	–	44 ± 2.49	697.53–916.51	769.91 ± 43.51	40.53–47.96	43.91 ± 1.83
	Thorax	25.6 ± 0.84	9.63 ± 1.78	118.08–366.39	261.28 ± 55.99	3.16–8.96	6.54 ± 1.25
	Abdomen and pelvis	26.93 ± 0.87	20.85 ± 2.46	498–974.91	737.28 ± 117.48	12.07–19.24	14.81 ± 1.84
Aquillion PRIME²⁰¹⁶	Head	–	35.57 ± 3.52	3228–7654	602.88 ± 57.78	35.87–43.97	40.03 ± 2.03
	Thorax	26 ± 1.24	7.76 ± 2.07	125.62–335.19	202.44 ± 40.28	3.7–8	5.18 ± 1.01
	Abdomen and pelvis	27.38 ± 0.86	12.17 ± 5.39	2329–5476	360.16 ± 82.77	4.8–11.4	7.55 ± 1.67
SOMATOM Definition AS+²⁰¹⁴	Head	–	53.18 ± 4.74	570.05–892.35	719.05 ± 75.41	37.36–48.03	42.58 ± 2.88
	Thorax	27.6 ± 1.36	7.58 ± 1.36	1424–3081	213.05 ± 42.4	3.12–9.02	5.49 ± 1.1
	Abdomen and pelvis	28.36 ± 0.89	12.43 ± 3.03	2356–624.99	432.76 ± 121.23	5.22–12.98	9.24 ± 2.46
SOMATOMDefinition AS+²⁰¹⁵	Head	–	62.1 ± 5.05	721.72–938.72	837.41 ± 53.25	44.25–51.02	47.99 ± 1.84
	Thorax	26.36 ± 0.92	5.16 ± 0.88	94.86–242.83	141.04 ± 32.1	2.25–4.98	3.49 ± 0.68
	Abdomen and pelvis	28.25 ± 0.79	10.46 ± 1.7	274.63–681.59	372.15 ± 92.4	5.71–11.79	7.34 ± 1.62
SOMATOM Definition AS+^2014*	Head	–	51.81 ± 6.95	583.91–7811	655.6 ± 44.06	41.82–49.82	46.64 ± 2.31
	Thorax	26.93 ± 1.43	7.77 ± 0.92	122–280.64	204.45 ± 37.28	4.04–7.75	5.52 ± 0.94
	Abdomen and pelvis	28.42 ± 0.92	15.45 ± 2.53	328.08–764.49	549.53 ± 108.4	7.43–16.27	11.49 ± 2
SOMATOM Definition AS+²⁰¹⁶	Head	–	52.72 ± 4.37	628.51–835.13	727.57 ± 52.74	41.46–49.38	44.98 ± 2.15
	Thorax	26.78 ± 1.62	8.74 ± 1.5	120.84–296.84	217.12 ± 44.4	3.5–8.85	6.1 ± 1.27
	Abdomen and pelvis	28.12 ± 0.97	12.17 ± 1.05	2865–517.88	376.01 ± 44.89	6.66–10.87	8.03 ± 0.77

Values are given as mean ± SD or range. Dash (–) indicates not applicable.

CTDI_vol, volume computed tomography dose index; DLP, dose length product; SSDE, size-specific dose estimate; WED, water equivalent diameter.

Data analysis

Descriptive data analysis was conducted, including mean, median, range, and standard deviation and 75th percentile. To establish the LDRLs, the third quartile (75th percentile) was rounded and compared with available national, local, and international data. These analyses were performed using Excel 2021 (Microsoft Corp., Redmond, WA, USA). Descriptive statistical analysis of the collected dosimetry parameters CTDI_vol and DLP was conducted for each scanner and each examination type. By using SPSS Statistics version 29 (IBM Corp., Armonk, NY, USA), an ANOVA test was conducted to look at the differences in radiation doses between scanners, with P < 0.05 indicating statistical significance.

The collected patient radiation dose for each body part and scan type were compared with LDRLs from Tonopki et al. (16) and NDRLs in Norway (17). To represent an international set of DRLs, this studýs results were also compared with data from Canada (2016 and 2022), the UK (2019), Australia (2021), France (2019), Switzerland (2020), and the USA (2017) (10,16,18 –22). It is important to note that there is no NDRL for the abdominal and pelvic examination in Norway.

Ethical perspectives

This retrospective quality assurance study included only dose data and scan parameters. No patient-sensitive data were registered and the included patients were not traceable. According to the guidelines from the Norwegian Centre for Research Data (NSD), it was concluded that approval from The Norwegian Agency for Shared Services in Education and Research (SIKT) was not necessary for this study.

Results

Table 2 shows the scan parameters for each scanner included in this study. All examinations utilized helical scanning. AEC was implemented on all examinations, except for the head CT on the S. Definition AS + ²⁰¹⁵, which uses fixed mAs. Iterative reconstruction (IR) algorithms specific to each manufacturer and model were used for all examinations except for the head examination from Discovery 690²⁰¹⁰. Instead, the GE Healthcare-specific recon types Plus and IQ Enhance were used.

Results of the statistical analysis presenting mean ± standard deviations for the WED, SSDE, CTDI_vol, and DLP values and ranges of all examinations on each scanner are reported in Table 3. Figs. 1–3 demonstrate the distribution of median CTDI_vol and DLP values from all scanners for each body part. The highest CTDI_vol and DLP values for the head examinations were obtained from the S. Definition AS + ²⁰¹⁵ scanner (Fig. 1). The Discovery 690²⁰¹⁰ had the highest CTDI_vol and DLP values for the thorax and the abdominal and pelvic examination (Figs. 2–3). The Aquillion PRIME²⁰¹⁶ had the lowest CTDI_vol and DLP, respectively, for the head examinations (Fig. 1), whereas the S. Definition AS + ²⁰¹⁵ had the lowest CTDI_vol and DLP for the thorax and abdominal and pelvic examinations (Fig. 2–3).

Fig 1.

Distribution of the median (a) CTDI_vol and (b) DLP values from all the scanners for the head examinations. The horizontal lines correspond to the national DRLs in Norway and the determined local DRLs. CTDI_vol, volume computed tomography dose index; DLP, dose length product; DRL, diagnostic reference level.

Fig 2.

Distribution of the median (a) CTDI_vol and (b) DLP values from all the scanners for the thorax examinations. The horizontal lines correspond to the national DRLs in Norway and the determined local DRLs. CTDI_vol, volume computed tomography dose index; DLP, dose length product; DRL, diagnostic reference level.

Fig 3.

Distribution of the median (a) CTDI_vol and (b) DLP values from all the scanners for the abdomen and pelvis examinations. The horizontal lines correspond to the national DRL in Norway and the determined local DRLs. CTDI_vol, volume computed tomography dose index; DLP, dose length product; DRL, diagnostic reference level.

Accordingly, the highest and lowest SSDE for the head (67.15 mGy and 32.06 mGy) were obtained from the S. Definition AS + ²⁰¹⁵ and the Aquillion Prime²⁰¹⁶, respectively. Similarly, the S. Definition AS + ²⁰¹⁵ and the Discovery 690²⁰¹⁰ scanners had the lowest and the highest SSDE for thorax and abdominal and pelvic examinations (4.28 vs 8.76 mGy and 11.4 vs 23.31 mGy, respectively) (Table 3).

The difference between the mean values of the dose distribution from identical scanner models at different locations (S. Definition AS + ^{2014, 2015, 2016}) was significantly different (P < 0.05) for all examinations. In addition, there is a statistically significant difference (P < 0.05) between the means of the dose distribution collected from each scanner for all examinations. The mean WED of the patients for the thorax protocol was found to be significantly smaller for the Discovery 690²⁰¹⁰ scanner compared to the rest of the scanners (P < 0.01). However, for the abdomen and pelvis protocol, no significant difference was found between the mean WED of the patients of the Discovery 690²⁰¹⁰ compared to that of the Aquillion Prime²⁰¹⁶ (P = 0.109). Among the Siemens scanners, patient size was found be significantly different for the thorax protocol (P < 0.01) even though there were no differences in patient size for the abdomen protocol (P > 0.53).

The registered 75th percentile LDRLs in this study for the CTDI_vol values for the head, thorax and abdominal and pelvic examinations were 46 mGy, 6 mGy, and 11 mGy, respectively. The corresponding DLP values were 762, 217, and 513 mGycm (Fig. 1–3). All scanners had median dose values lower than the Norwegian NDRLs for examinations of the head and thorax. The abdomen and pelvis LDRLs were also lower than the local and international DRLs, except for the Discovery 690²⁰¹⁰.

Fig. 4 shows a comparison between NDRLs and LDRLs in Norway with published international data. It should be noted that there are no established NDRLs for the abdominal and pelvic examination in Norway. The Norwegian NDRLs for the head and thorax fall within the same range as the international data. The LDRLs for all examinations in this study are lower than both the national and international values. However, the CTDI_vol (head) and DLP (abdomen and pelvis) of the Canadian LDRL from the study by Tonopki et al. (16) were slightly lower than the values obtained from this study (44 vs. 46 mGy and 476 vs. 513 mGycm, respectively).

Fig 4.

Comparison of the local DRLs determined in the current study with the local, national, and international published data (1 –8). Note, that there has not been established a DRL for the abdomen and pelvis examination in Norway. DRL, diagnostic reference level.

Discussion

This study has established LDRLs from six Norwegian private diagnostic facilities, for frequently performed CT protocols, including routine non-enhanced head scans, contrast-enhanced CT of the thorax, and contrast-enhanced CT of the abdomen and pelvis. Registered doses for each body part and scan type were compared with local, national, and international DRLs to identify the need for protocol optimization in the involved facilities. The LDRLs included in this study were lower than the Norwegian NDRLs for CT examinations of the head and thorax (Fig. 1 and 2). The LDRLs for almost all examinations in this study were also lower than the international values (Fig. 4).

A comparison between scanners in this study demonstrated wide variations in the dose indicators confirmed by the statistical analysis (P < 0.05). This is likely due to differences in patient size (WED), scanners type, and exposure parameters, especially AEC and IR. All three scanner manufacturers use AEC, which automatically adapts tube current to patient size to achieve a specified image quality or level of image noise (8). IR is an algorithm that reconstructs an image through a series of continual iterations, each one improving the quality of the image (23). The use of IR produces CT images with lower noise levels and could lead to substantial dose reduction, which depends on the algorithm type (2,13). The difference between doses from identical scanner models (S. Definition AS + ^{2014, 2015, 2016)}) at different locations was also significantly significant (P < 0.05) for all examinations. This might be because of the difference in scan parameters or patient size. However, no significant differences (P > 0.53) were observed between mean WED of the patients for the abdomen and pelvis protocol for all Siemens scanners.

Results from the Siemens scanners point to the fact that a combination of AEC (lower reference mAs), use of CarekV, and higher levels of IR contributes to the lower doses. Thus, all protocols with CarekV resulted in lower doses compared with protocols with semi or no CarekV. CarekV automatically selects the optimal kV settings to save dose. In CARE kV semi mode, a user-defined kV is implemented, and the mAs is modulated accordingly, to provide the contrast-to-noise ratio (CNR) defined by the CARE kV reference kV, mAs, and scan type. This can result in different choices of kV for patients of the same size (16,24). Other researchers have also shown that CarekV gives more dose reduction than CarekV semi (24). The S. Definition AS + ²⁰¹⁵ had the highest CTDI_vol and DLP values for the head examinations (47.97 and 835.06 mGycm, respectively) even though it had the highest pitch (Fig. 1). This is caused by the fact that it is the only scanner using fixed mAs and the lowest level of IR out of all the other scanners (2,11 –13). It is also worth mentioning that, even though the CTDI_vol of the S. Defintion²⁰¹⁴* scanner was found to be higher than that of the S. Defintion²⁰¹⁴ (46.8 vs. 42.6 mGy), the DLP of the S. Defintion²⁰¹⁴* was lower than that of the S. Defintion²⁰¹⁴ scanner (655.6 vs 719.1 mGycm), indicating the use of a shorter scan length. The scanner operator determines the scan length, and a shorter scan length will decrease the DLP value (2,10). This suggests that the S. Defintion²⁰¹⁴ scanner may require scan length optimization for the head protocol.

For the thorax and abdominal and pelvic examinations, the Discovery 690²⁰¹⁰ had the highest doses (6.52/256.89 mGycm and 14.8/737.28 mGycm, respectively) although the mean WED of the patients for these protocols were significantly smaller than that of the Siemens scanners (P < 0.01). This might be due the fact that the scanner is from 2010, and it does not have the latest advancements in CT technology to enable substantial dose reduction and the AEC settings used are also high (8).

The 75th percentile LDRLs for all examinations in this study were lower than the Norwegian and international values as shown in Fig. 4. Nevertheless, the DLP of the abdomen and pelvis protocol from Canadian LDRL (Tonopki et al. (16)) was lower compared to the current study (476 vs 509 mGycm). When comparing the abdomen and pelvis protocols from this study with the Canadian protocols by Tonopki et al. (16), some similarities are evident. However, the scanners used in this study appear to have higher reference mAs than those in Tonopki et al. (16). For example, the Discovery 690²⁰¹⁰ has a higher median dose than the rest of the scanners for the abdomen and pelvis. Further investigation is needed to determine whether lower mAs and higher pitch can be used with this scanner to optimize the dose while maintaining acceptable image quality.

A limitation of this study is that it does not collect information about image quality (10,25). However, all the included examinations met the institution's quality standards and were clinically accepted.

In conclusion, it was evident that the LDRLs from this study were lower overall, for both national and international DRLs. Variations in median dose values among scanners of the same make and model suggest potential for protocol optimization in some facilities. There also seems to be an opportunity to establish updated NDRLs in Norway for head and thorax examinations, as well as a new NDRL for the abdominal and pelvic examinations.

Footnotes

Acknowledgments

The authors are grateful for help from the facilities involved in this study.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Frida Gravdahl Helgesen

Safora Johansen

References

Ninsiima

Boadu

Inkoom

, et al. Dose-image quality assessment for optimization of adult chest computed tomography examinations: a phantom study. Uganda/Ghana: ICRP, 2021.

Vañó

Miller

Martin

, et al. ICRP Publication 135: diagnostic reference levels in medical imaging. ICRP 2017;46:1–144.

Elnour

Barakat

Tamam

, et al. Estimation of radiation risk and establishment of diagnostic reference levels for patients undergoing computed tomography chest-abdomen-pelvic examinations in Sudan. Radiat Prot Dosimetry 2021;196:104–109.

Foley

McEntee

Rainford

. Establishment of CT diagnostic reference levels in Ireland. Br J Radiol 2012;85:1390–1397.

Wildberger

Mahnken

Schmitz-Rode

, et al. Individually adapted examination protocols for reduction of radiation exposure in chest CT. Invest Radiol 2001;36:604–611.

Donnelly

Emery

Brody

, et al. Minimizing radiation dose for pediatric body applications of single-detector helical CT: strategies at a large Children’s Hospital. AJR Am J Roentgenol 2001;176:303–306.

Frush

Soden

Frush

, et al. Improved pediatric multidetector body CT using a size-based color-coded format. AJR Am J Roentgenol 2002;178:721–726.

Lee

Goo

, et al. Radiation dose modulation techniques in the multidetector CT era: from basics to practice. Radiographics 2008;28:1451–1459.

Martin

Vano

. Diagnostic reference levels and optimisation in radiology: where do we go from here? J Radiol Prot 2018;38:E1–E4.

10.

Kanal

Butler

Sengupta

, et al. U.S. diagnostic reference levels and achievable doses for 10 adult CT examinations. Radiology 2017;284:120–133.

11.

Awad

Karout

Arnous

, et al. A systematic review on the current status of adult diagnostic reference levels in head, chest and abdominopelvic computed tomography. J Radiol Prot 2020;40:71–98.

12.

McCollough

Bakalyar

Bostani

, et al. Use of water equivalent diameter for calculating patient size and size-specific dose estimates (SSDE) in CT: the report of AAPM task group 220. AAPM Rep 2014;2014:6–23.

13.

Boos

Thomas

Appel

, et al. Institutional computed tomography diagnostic reference levels based on water-equivalent diameter and size-specific dose estimates. J Radiol Prot 2018;38:536–548.

14.

Sutton

Worrall

Sexton

, et al. The influence of patient size on the overall uncertainty in radiographic dose audit. J Radiol Prot 2021;41: 539–551.

15.

Leng

Shiung

Duan

, et al. Size-specific dose estimates for chest, abdominal, and pelvic CT: effect of intrapatient variability in water-equivalent diameter. Radiology 2015;276:184–190.

16.

Tonkopi E, Wikan EJ, Hovland TO, et al. A survey of local diagnostic reference levels for the head, thorax, abdomen and pelvis computed tomography in Norway and Canada. Acta Radiol Open 2022;11(10).

17.

Directorate For Radiation Protection and Nuclear Safety. Representative doser i Norge - 2017. Strålevernsrapport 2018:3. Published 22 Febuary 2018. Available at: https://dsa.no/publikasjoner/stralevernrapport-3-2018-representative-doser-i-norge-2017/StralevernRapport_2017-3_Representative%20doser%202017.pdf (accessed 21 January 2024).

18.

Health Canada. Canadian computed tomography survey: national diagnostic reference levels. Published 14 June 2016. Available at: https://www.canada.ca/en/health-canada/services/publications/health-risks-safety/canadian-computed-tomography-survey-national-diagnostic-reference-levels.html (accessed 21 January 2024).

19.

Public Health England. National Diagnostic Reference Levels (NDRLs) from 19 August 2019. Available at: https://www.gov.uk/government/publications/diagnostic-radiology-national-diagnostic-reference-levels-ndrls/ndrl. (accessed 21 January 2024).

20.

Australian Radiation Protection and Nuclear Safety Agency. Current Australian national diagnostic reference levels for multi detector computed tomography from March 2021. Available at: https://www.arpansa.gov.au/research-and-expertise/surveys/national-diagnostic-reference-level-service/current-australian-drls/mdct (accessed 21 January 2024).

21.

IRSN. Analyse des donńees relatives `a la mise `a jour des niveaux de ŕef´erence diagnostiques en radiologie et en m´edecine nucĺeaire. year. France: IRSN, 2020.

22.

Aberle

Ryckx

Treier

, et al. Update of national diagnostic reference levels for adult CT in Switzerland and assessment of radiation dose reduction since 2010. Eur Radiol 2020;30:1690–1700.

23.

ICRU. ICRU Report No. 87: Radiation dose and image-quality assessment in computed tomography. J ICRU 2012;1:1–149.

24.

Bebbington

Jørgensen

Dupont

, et al. Validation of CARE kV automated tube voltage selection for PET-CT: PET quantification and CT radiation dose reduction in phantoms. EJNMMI Phys 2021;8:29.

25.

Kanal

Butler

Chatfield

, et al. U.S. Diagnostic reference levels and achievable doses for 10 pediatric CT examinations. Radiology 2022;302:164–174.