Effect of iterative reconstruction algorithm levels on noise index and figure-of-merit in CT pulmonary angiography examinations

Abstract

PURPOSE:

To evaluate the influence of iterative reconstruction (IR) levels on Computed Tomography (CT) image quality and to establish Figure of Merit (FOM) value for CT Pulmonary Angiography (CTPA) examinations.

METHODS:

Images of 31 adult patients who underwent CTPA examinations in our institution from March to April 2019 were retrospectively collected. Other data, such as scanning parameters, radiation dose and body habitus information from the subjects were also recorded. Six different levels of IR were applied to the volume data of the subjects. Five circles of the region of interest (ROI) were drawn in five different arteries namely, pulmonary trunk, right pulmonary artery, left pulmonary artery, ascending aorta and descending aorta. The mean Signal-to-noise ratio (SNR) was obtained, and the FOM was calculated in a fraction of the SNR² divided by volume-weighted CT dose index (CTDI_vol) and SNR² divided by the size-specific dose estimates (SSDE).

RESULTS:

Overall, we observed that the mean value of CTDI_vol and SSDE were 13.79±7.72 mGy and 17.25±8.92 mGy, respectively. Notably, SNR values significantly increase with increase of the IR level (p < 0.05). There are also significant differences (p < 0.05) in the FOM for both SNR²/SSDE and SNR²/CTDI_vol attained in different IR levels.

CONCLUSION:

We successfully evaluate the value of radiation dose and image quality performance and set up a figure of merit for both parameters to further verify scanning protocols by radiology personnel.

Keywords

Signal-to-noise ratio figure-of-merit CT pulmonary angiography pulmonary embolism iterative reconstruction algorithm

1 Introduction

Computed Tomography Pulmonary Angiography (CTPA) examination is one of the standard imaging techniques for detecting Pulmonary Embolism (PE). PE is a condition when a clot blockage inside the pulmonary arteries and interrupt the circulation of the blood in a respiratory system [1]. Although CTPA has become a prominent imaging procedure for detecting PE, the false-negative indications from the techniques are inevitable. Therefore, the frequency of the examinations has increased and could contribute up to 10 mSv of effective dose from one single examination [2, 3]. As a result, the demands of the CTPA examination could increase the effect of radiation-induced risk to the patients [4, 5].

The most prominent indicator for representing patient radiation exposure in CT examination is volume-weighted CT Dose Index (CTDI_vol). 32 cm diameter size of an ellipsoid phantom is used as a for estimating CTDI_vol and considered as a surrogate of the human body [6]. However, CTDI_vol has a significant error since the patient size is not as fixed as a standard phantom. A recent study has shown that patient dose is inversely proportional to the size of the patient since the patient’s effective diameter is mostly less than 32 cm [7, 8]. As the size of the patient decreases from a standard phantom, the radiation exposure of the patient also increased. Hence, to correspond with the size of patients, the Size-specific Dose Estimates (SSDE) was introduced by the American Physicians Association (AAPM) in 2011 [9, 10]. SSDE is using patient-conversion factors as a substitute for patient size to estimate more precise exposure rather than CTDI_vol [11].

The diagnostic performance needs to balance with the reduction of doses as prompted by the ALARA principle [12]. One of the critical features of optimisation is the improvement of image quality technology that allows for removing the artefacts and suppressing the noise on the images [13, 14]. Thus, current CT has introduced an impressive software package which is iterative reconstruction (IR) algorithm. IR required low spectrum energy to attain the same diagnostic performance by the previous version of the image reconstruction; filter-back projection (FBP) [15]. More advances IR model like Model-based iterative reconstruction (MBIR) promotes noticeable noise reduction simultaneously improved the SNR values.

Moreover, it permits additional edge-preserving regularization to improve contrast resolution of the CT images [16, 17]. However, recent work showed that the tube potential reduction (CARE kV scanning) alone would cause increased noise and reduced the SNR if not considering the IR algorithm adjustment [18]. It may lead to misinterpretation of the diagnosis and should be applied cautiously.

The Figure of Merit (FOM) is a well-known measurement to evaluate both the image quality and radiation exposure equally [19 –21]. Previous works were reported the effective dose (ED) as a different approach to determine the dose descriptor for FOM calculation beside the SSDE and CTDI_vol [2 , 22]. Nevertheless, none of the prior studies employs SSDE as a function of radiation dose to determine the FOM that proven more accurate for estimating the radiation dose. Therefore, the study aims to assess the implication of the various levels of iterative reconstruction algorithm on image quality performance and to establish the index of the FOM based on the comparison between the CTDI_vol and SSDE approach.

2 Materials and methods

2.1 Patient population

Thirty-one adults (17 males and 14 females) who were being clinically indicated for PE underwent CTPA examinations were collected retrospectively. Mean age was 35.9 years with a range from 18– 77 years for male and 40.9 years with a range from 18– 67 years for female patients. This study was approved by the Medical Research and Ethics Committee (MREC), Ministry of Health Malaysia (MOH) with the approval number: NMRR-18-3088-44138. The exclusion criteria in this study are the case with a motion artefact and faint contrast conditions.

2.2 CT acquisition and protocols

The CTPA procedure was performed by using a 128 multi-detector CT (Philips Brilliance) scanner which installed at the Radiology Department, Hospital Kuala Lumpur Malaysia. The subjects were injected with 40 to 70 ml of iodinated contrast medium (370 mgI/ml concentration), followed by 50 ml saline contrast at a flow rate of 5 ml/s. The bolus tracker method was performed by placing the region of interest (ROI) on the main pulmonary trunk. After reached a threshold of 70 HU, the scanning was started with a range of 7– 14 seconds depending on body habitus. The automated tube current modulation (ATCM) function was enabled for the protocol to compensate the tube current of the patient along the z-axis. For post-processing, the images were reconstructed using the available IR algorithm (iDose⁴) with 1 mm slice thickness and 512×512 matrix size. All radiology personnel involved in the procedure were well experienced for at least three years. Table 1 summarized scanning acquisition parameter of CVTPA examinations.

Table 1
Data on scanning acquisition in CTPA examination

Scanning Parameter Values

Tube Voltage (Kv) 100 and 120 kV

Tube Current (mAs)* 184.50±81.87

Scan Range (mm)* 277.44±88.03

Pitch Factor 0.798

Beam Collimation (mm) 0.625×40

Rotation Time (s) 0.50

Scanning Parameter	Values
Tube Voltage (Kv)	100 and 120 kV
Tube Current (mAs)*	184.50±81.87
Scan Range (mm)*	277.44±88.03
Pitch Factor	0.798
Beam Collimation (mm)	0.625×40
Rotation Time (s)	0.50

Fig.1

Body’s effective diameter measurement at the mid-slice of the 3D CT images.

2.3 Radiation dose measurement

The calculation of SSDE was based on the derivation of CTDI_vol and effective diameter with the coefficient obtained from a documented report supplied by the AAPM (Report 204). The effective diameter of human body was determined, as illustrated in Fig. 1, using the following equation: $Effective Diameter (cm) = \sqrt{AP \times LAT}$ (1)

Finally, the SSDE was estimated by using the following equation: $SSDE = {CTDI}_{VOL} \times f_{size}$ (2) where the CTDI_vol was acquired directly from the dose report in scanner system, and f_size is a conversion factor according to the effective diameter as obtained by the AAPM document. The CTDI_vol was defined by software calculation in the scanner system based on a 32 cm standard phantom. It is highly dependent on the detector collimation used and table movement per single rotation of the x-ray tube. Each effective diameter measured was converted into the coefficient value (f_size) represent the normalise dose coefficient for the 32 cm PMMA CTDI_vol phantom.

2.4 Image quality and FOM assessment

Fig.2

Placement of the ROI for the SNR calculation.

Fig.3

Comparison of selected slice of pulmonary artery images at different IR level (a) iDose⁴ level 1, (b) iDose⁴ level 2, (c) iDose⁴ level 3, (d) iDose⁴ level 4, (e) iDose⁴ level 5 and (f) iDose⁴ level 6.

Image quality performance of different IR levels was indicated by quantifying the Signal-to-Noise ratio (SNR) value. This assessment was performed by a medical physicist with more than five years of experience by using mini PACS (picture archiving and communication system) workstation. Each retrospective image using default IR setting (iDose⁴ level 4) was reconstructed again to different IR levels available in the scanner system (iDose⁴ level 1, 2, 3, 5 and 6). 5 circular shape with size the average of 150 mm² were drawn as ROI on the surrounding pulmonary artery respectively on the main pulmonary trunk, right pulmonary artery, left pulmonary artery, ascending aorta and descending aorta as shown in Fig. 2. If PE was present, ROI circles were meticulously drawn without incorporating embolic material. The ROI will be drawn fit to the artery area if the size of the artery decreased. CT number and standard deviation (SD) value from the ROI were recorded for the analysis purpose for different IR level as shown in Fig. 3. The SNR of each artery was calculated by using the following formula [22]: $SNR = \frac{{HU}_{artery}}{{SD}_{artery}}$ (3) where HU_artery and SD_artery, are a CT number value and noise value of selected pulmonary arteries. The FOM corresponds to radiation dose, CTDI_vol and SSDE for each different IR levels, was estimated using the following equation [21]. $FOM = \frac{I^{2}}{D}$ (4) where I is the quantitative value of image quality (SNR), and D is a radiation dose value (CTDI_vol or SSDE) in mGy.

2.5 Statistical data analysis

Data were evaluated by using Statistical Package for Social Science (SPSS) v25.0 (Chicago, Illinois, USA). All variables were expressed as a mean value and presented descriptively. Shapiro-Wilk test was used to determine the normality of the data. The data were compared statistically using a one-way ANOVA test and Kruskal-Wallis test with a p-value set at <0.05.

3 Results

3.1 Radiation measurement and diagnostic performances

Table 2 tabulates the radiation dose of different categories of tube voltage and the body’s effective diameter. Subjects were categorised into three distinct groups concerning their body effective diameter: Group 1 (20– 24 cm), Group 2 (24– 28) and Group 3 (28– 38 cm) based on descriptive analysis. Considerable variations of CTDI_vol were observed with different tube potential and effective diameter by 180% and 45%, respectively. Besides, there were showed a significant difference in the comparison of CTDI_vol and SSDE (p < 0.05). Table 3 and Fig. 4 presents the mean CT- number and mean noise for all levels. Notably, the percentage reduction between 1% to 13% of were observed due to the increase of the IR level. Interestingly, the mean CT- number was slightly decreased along with the rise of the iDose⁴ level. A noticeable difference of the mean SNR was obtained depending on IR level in CT scanner with up to 41%.

Table 2
A comparison of CTDI_vol and SSDE value with different tube voltage and the body’s effective diameter

Tube Potential/ 100 kV 120 kV

Body Effective (20– 24) (25– 28) (29– 38) (20– 24) (25– 28) (29– 38)

Diameter Group (cm) Group 1 Group 2 Group 3 Group 1 Group 2 Group 3

Number of subject (n) 5 7 3 2 7 7

Mean CTDI_vol (mGy)* 5.44±1.21 6.22±1.01 7.89±2.07 13.95±4.33 17.65±3.93 19.89±4.81

Mean SSDE (mGy)* 8.39±1.03 8.73±1.15 10.01±1.98 20.53±4.13 22.19±3.71 24.33±4.92

p-value <0.05 <0.05 <0.05 <0.05 <0.05 <0.05

Tube Potential/	100 kV	120 kV
Number of subject (n)	5	7	3	2	7	7
Mean CTDI_vol (mGy)*	5.44±1.21	6.22±1.01	7.89±2.07	13.95±4.33	17.65±3.93	19.89±4.81
Mean SSDE (mGy)*	8.39±1.03	8.73±1.15	10.01±1.98	20.53±4.13	22.19±3.71	24.33±4.92
p-value	<0.05	<0.05	<0.05	<0.05	<0.05	<0.05

Table 3

A comparison of all variable values obtained from different iterative reconstruction levels

Variable/	iDose⁴ level						p-value
IR levels		1	2	3	4	5	6
Mean CT-Number	MPA	373.59±152.24	373.56±152.23	373.47±152.09	373.41±152.07	373.40±152.04	373.29±151.96	NS
	RPA	350.44±132.38	350.31±132.36	350.29±132.48	349.87±132.28	349.86±132.28	349.23±116.52	NS
	LPA	352.70±116.43	352.51±116.50	352.35±116.40	352.18±116.49	352.08±116.39	352.12±132.50	NS
	AA	262.49±113.55	262.49±113.55	262.51±113.52	262.54±113.56	262.54±113.58	262.54±113.57	NS
	DA	246.09±106.03	245.93±105.99	245.74±105.86	245.70±105.82	245.56±105.79	245.38±105.64	NS
Mean Noise	MPA	26.34±5.62	25.07±5.80	23.74±5.87	22.30±6.06	20.75±6.49	19.03±6.02	<0.05
	RPA	62.29±17.41	61.36±18.87	59.86±19.13	59.16±19.86	57.81±18.72	56.60±10.13	<0.05
	LPA	44.43±12.41	43.52±12.14	42.38±13.26	41.19±11.96	39.96±12.13	38.70±11.97	<0.05
	AA	28.55±10.13	27.27±9.50	25.88±9.41	24.56±8.97	23.10±8.85	21.48±8.94	<0.05
	DA	28.09±8.13	26.69±8.24	25.09±8.15	23.61±8.62	21.90±8.64	20.06±8.86	<0.05
Mean SNR	MPA	14.98±7.37	15.80±7.38	16.79±7.40	17.99±7.42	19.49±7.45	21.54±7.46	<0.05
	RPA	7.22±3.47	7.50±3.80	7.85±4.22	8.20±4.73	8.69±3.39	9.23±6.46	<0.05
	LPA	9.46±3.49	9.78±3.73	10.19±4.05	10.68±4.45	11.26±4.99	11.44±5.68	<0.05
	AA	8.89±3.84	9.14±4.10	9.85±4.31	10.50±4.75	11.22±5.17	12.46±5.74	<0.05
	DA	8.84±3.81	9.20±4.13	9.89±4.43	10.62±4.75	11.46±5.41	12.08±6.25	<0.05

*MPA = main pulmonary artery, RPA = Right pulmonary artery, LPA = Left pulmonary artery, AA = Ascending aorta, DA = Descending aorta. **NS: not significant.

3.2 FOM calculation

The results of FOM in five different areas of the pulmonary arteries are listed in Table 4. It appears that all parameters are increased with increasing IR level with 70% and differ significantly (p < 0.05). Besides, a significant reduction of the FOM is shown for SNR²/SSDE compared to SNR²/CTDI_vol values and with increasing body effective diameter sizes as tabulated in Table 5.

Fig.4

Mean CT-number (top left), mean noise (top right), mean SNR (bottom left) in different iDose⁴ level for five pulmonary artery area and comparison between two FOM approach (bottom right) of 3 groups with differently effective diameter (cm) ranges in CTPA examination.

Table 4

The figure of merit (FOM) for the relationship between SNR/CTDI_vol and SNR/SSDE

Variable	iDose⁴ level						p-value
		1	2	3	4	5	6
FOM SNR²/CTDI_vol	MPA	15.60±6.06	17.28±6.80	19.30±6.62	22.10±6.13	25.71±6.91	31.10±7.38	<0.05
	RPA	3.95±1.07	4.24±1.30	4.59±1.87	4.92±2.32	5.54±2.69	6.22±2.74	<0.05
	LPA	7.93±2.58	8.51±2.74	9.24±2.95	10.31±3.72	11.61±3.82	13.30±3.74	<0.05
	AA	5.92±3.57	6.56±4.10	7.33±4.15	8.44±4.78	9.68±4.71	11.68±5.78	<0.05
	DA	5.82±2.15	6.47±3.17	7.29±3.60	8.60±4.64	9.88±5.32	12.07±6.80	<0.05
FOM SNR²/SSDE	MPA	11.22±6.16	12.43±8.09	13.90±8.80	15.95±9.98	18.60±9.79	22.58±10.92	<0.05
	RPA	2.90±1.06	3.11±1.92	3.37±2.03	3.63±2.74	4.08±3.12	4.57±3.64	<0.05
	LPA	5.78±2.70	6.21±3.57	6.75±3.43	7.54±4.13	8.49±4.41	9.75±4.61	<0.05
	AA	4.54±2.53	5.04±2.89	5.64±3.71	6.50±3.90	7.45±4.52	9.01±4.78	<0.05
	DA	4.44±1.45	4.94±1.90	5.57±2.65	6.56±4.12	7.55±4.94	9.20±5.06	<0.05

Table 5

Comparison between two Figure of Merit (FOM) in different body effective diameter

Variable	Body Effective Diameter Group (cm)	iDose⁴ level						p-value
		1	2	3	4	5	6
FOM SNR²/CTDI_vol	(20– 24) Group 1	27.55±12.74	29.87±14.64	33.18±16.79	37.03±18.31	40.90±19.54	49.67±21.92	<0.05
	(25– 28) Group 2	11.08±6.92	12.23±7.98	13.72±9.58	15.47±10.83	17.86±11.98	21.30±13.90	<0.05
	(29– 38) Group 3	8.64±4.75	9.18±6.42	11.14±7.31	12.81±8.58	14.97±9.30	18.56±11.33	<0.05
FOM SNR²/SSDE	(20– 24) Group 1	19.72±6.45	21.38±6.08	23.74±6.88	26.48±7.94	29.23±9.50	35.50±11.03	<0.05
	(25– 28) Group 2	8.64±5.68	9.52±5.67	10.66±6.00	12.02±6.83	13.82±7.97	16.44±9.52	<0.05
	(29– 38) Group 3	6.79±3.67	7.23±4.13	8.76±4.75	10.07±5.64	11.78±6.84	14.62±7.96	<0.05

4 Discussion

This study identified the relationship of iterative reconstruction levels with image quality performance (SNR) and FOM approach. The SSDE, which is body sizes-dependent, reveals the inaccurate estimation of the CTDI_vol. The CT-system has estimated the values with a constant 32 cm body phantom as a coefficient metric independent with body sizes [23]. As reported by the AAPM, the effective diameter provides a unique individual body factor into the SSDE calculation [20]. As expected, this study was parallel with previous research that both CTDI_vol and SSDEs increased as the body effective diameter increased in ATCM mode [2 , 24].

The increasing IR level causes a significant improvement of both the SNR and FOM values. A previous study showed that increasing the IR contributes to the reduction of the noise magnitude [25]. However, image texture alterations were produced that resulting in the degradation of CT-images with too high IR [26]. Thus, the optimal level of IR needs to apply to overcome this matter wisely. If selecting lower IR level (e.g., the iDose⁴ level 1) can cause the undesired level of noise meanwhile with higher IR level (e.g., the iDose⁴ level 6) will create a “plastic” or too smooth appearance in CT images that obstruct diagnostic information [20].

IR algorithm adjustment is a crucial role for increasing diagnostic performance of CT examination. Previous research reported that the IR algorithm contributes to superior SNR value in comparison to filter-back projection (FBP) [27]. Also, this approach helps to display tiny vessels and other small structures such as a segmental artery, small bronchial walls and pleura [28]. This study showed that using the combination of IR algorithm and ATCM; one could obtain similar image quality with decreasing the radiation dose [29]. Hence, it provided a baseline to propose better low dose protocols in future CTPA examination. Meanwhile, lower-energy techniques will result in increased image noise via increased quantum mottle, the improved vascular attenuation at low kilovoltage settings may result in comparable SNR when compared to higher kV settings [30]. Instead, with high radiation exposure setting for the CT-examination will increase SNR values significantly with the reduction of noise. Still, there is the necessary caveat; it does not mean the additional diagnostic information shall be produced automatically with reduction of noise [31]. Thus, the dose adjustment must be dependent on the type of indication in CT-examination needed.

A notable reduction of the FOM value were found in greater body size or Body Mass Index (BMI) [19, 21]. The increase of radiation exposure with higher tube current will simultaneously affect the FOM values, which is a sum fraction of image quality over radiation exposure. According to some research report, increasing the tube current for the larger size of the patient is essential; it can provide better noise reduction and SNR level in CT images [19, 31]. The reduction patterns were expected due to increasing the tube current with and leads to a higher radiation dose. Multiple studies have reported reduced dose exposure, and improved image quality after changing the FBP to IR [20 , 23].

There were some limitations to this study. This study is covered only a small, retrospective nature of the patient cohort and depending only on the objective measurement. Further investigation should be detailed with the subjective image measurement by experiences imaging specialist to identify the relationship between both measures. Besides, this study was assessed only a single manufacturer’s scanner that not suitably represent with a different configuration of the IR algorithm in the inter-scanner types. Lastly, the best way for the optimisation process is to expand this work with protocol adjustment by the prospective patient to establish more appropriate protocol.

5 Conclusion

In this study, we successfully attain the comparable values between different dose descriptor, CTDI_vol and SSDE from CTPA examinations. The study results demonstrate that the SNR value increases along with the increase of the IR algorithm levels. The FOM values depend on the types of doses as denominator and body sizes. The finding from this study provides useful information on the iterative reconstruction algorithm, which would be essential for optimization in the CTPA examinations.

Footnotes

Acknowledgments

The authors wish the gratitude to the support from the radiology team from Hospital Kuala Lumpur. The author also wishes to acknowledge support from Geran Putra IPM of Universiti Putra Malaysia with the project no. GP/IPM/9619800.

References

Heit

J.A.

, Silverstein

M.D.

, Mohr

D.N.

, et al., Risk factors for deep vein thrombosis and pulmonary embolism, Arch Intern Med 160 (2000), 809–815.

Sauter

, Koehler

, Brendel

, et al., CT pulmonary angiography: dose reduction via a next generation iterative reconstruction algorithm, Acta Radiol 60 (2019), 478–487.

Karim

M.K.A.

, Sabarudin

, Muhammad

N.A.

, Ng

K.H.

A comparative study of radiation doses between phantom and patients via CT angiography of the intra-/extra-cranial, pulmonary, and abdominal/pelvic arteries, Radiol Phys Technol 12 (2019), 374–381.

Karim

M.K.A.

, Hashim

, Sabarudin

, et al., Evaluating organ dose and radiation risk of routine CT examinations in Johor Malaysia, Sains Malaysiana 45 (2016), 567–573.

Halid

, Karim

M.K.A.

, Sabarudin

, Bakar

K.A.

and Shariff

N.D.

, Assessment of Lifetime Attributable Risk of Stomach and Colon Cancer During Abdominal CT Examinations Based on Monte Carlo Simulation, In 6th International Conference on the Development of Biomedical Engineering in Vietnam (BME6), vol. 63, p. 455. Springer, 2017.

Daudelin

, Medich

, Andrabi

S.Y.

, Martel

Comparison of methods to estimate water-equivalent diameter for calculation of patient dose, J Appl Clin Med Phys 19 (2018), 718–723.

Anam

, Haryanto

, Widita

, et al., Automated calculation of water-equivalent diameter (DW) based on AAPM Task Group 220, J. Appl Clin Med Phys 17 (2016), 320–333.

Hashim

, Karim

M.K.A.

, Bakar

K.A.

, et al., Evaluation of organ doses and specific k effective dose of 64-slice CT thorax examination using an adult anthropomorphic phantom, Radiat Phys Chem 126 (2016), 14–20.

Burton

C.S.

, Szczykutowicz

T.P.

Evaluation of AAPM Reports 204 and 220: Estimation of effective diameter, water-equivalent diameter, and ellipticity ratios for chest, abdomen, pelvis, and head CT scans, J Appl Clin Med Phys 19 (2018), 228–238.

10.

Murphy

K.P.

, Crush

, O’Neill

S.B.

, et al., Feasibility of low-dose CT with model-based iterative image reconstruction in follow-up of patients with testicular cancer, Eur J Radiol Open 3 (2016), 38–45.

11.

Pourjabbar

Size-specific dose estimates: Localiser or transverse abdominal computed tomography images?, World J Radiol 6 (2014), 210–217.

12.

Karim

M.K.A.

, Hashim

, Bradley

D.A.

, et al., Radiation doses from computed tomography practice in Johor Bahru, Malaysia, Radiat Phys Chem 121 (2016), 69–74.

13.

Karim

M.K.A.

, Rahim

N.A.A.

, Matsubara

, et al., The effectiveness of bismuth breast shielding with protocol optimisation in CT Thorax examination, J Xray Sci Technol 27 (2019), 139–147.

14.

Isa

I.N.

, Rahmat

S.M.S.

, Dom

S.M.

, et al., The effects of mis-centering on radiation dose during CT head examination: A phantom study, J Xray Sci Technol 27 (2019), 631–639.

15.

Kalender

W.A.

Dose in x-ray computed tomography, Phys Med Biol 59 (2014), R129–R150.

16.

Desai

G.S.

, Thabet

, Elias

A.Y.A.

, Sahani

D.V.

Comparative assessment of three image reconstruction techniques for image quality and radiation dose in patients undergoing abdominopelvic multidetector CT examinations, Br J Radiol 86 (2013), 20120161.

17.

Ang

W.C.

, Hashim

, Khalis

, et al., Adaptive iterative dose reduction (AIDR) 3D in low dose CT abdomen-pelvic: Effects on image quality and radiation exposure, J Phys Conf Ser 851 (2017), 012006.

18.

Yang

, Li

Z.L.

, Gao

, et al., Image quality evaluation for CARE kV technique combined with iterative reconstruction for chest computed tomography scanning, Med (United States) 96 (2017), e6175.

19.

Kim

H.G.

, Lee

H.J.

, Lee

S.K.

, et al., Head CT: Image quality improvement with ASIR-V using a reduced radiation dose protocol for children, Eur Radiol 27 (2017), 3609–3617.

20.

Kligerman

, Mehta

, Farnadesh

, et al., Use of a hybrid iterative reconstruction technique to reduce image noise and improve image quality in obese patients undergoing computed tomographic pulmonary angiography, J Thorac Imaging 28 (2013), 49–59.

21.

Leyendecker

, Faucher

, Labani

, et al., Prospective evaluation of ultra-low-dose contrast-enhanced 100-kV abdominal computed tomography with tin filter: effect on radiation dose reduction and image quality with a third-generation dual-source CT system, Eur Radiol 29 (2019), 2107–2116.

22.

Nania

, Weir

, et al., CTPA protocol optimisation audit: challenges of dose reduction with maintained image quality, 320.e, Clin Radiol 73 (2018), 1–8.

23.

Laqmani

, Regier

, Veldhoen

, et al., Improved image quality and low radiation dose with hybrid iterative reconstruction with 80 kV CT pulmonary angiography, Eur J Radiol 83 (2014), 1962–1969.

24.

Sabel

B.O.

, Buric

, Karara

, et al., High-pitch CT pulmonary angiography in third generation dual-source CT: Image quality in an unselected patient population, PLoS One 11 (2016), E0146949.

25.

Verdun

F.R.

, Racine

, Ott

J.G.

, et al., Image quality in CT: From physical measurements to model observers, Phys Medica 31 (2015), 823–843.

26.

Itatani

, Oda

, Utsunomiya

, et al., Reduction in radiation and contrast medium dose via optimisation of low-kilovoltage CT protocols using a hybrid iterative reconstruction algorithm at 256-slice body CT: Phantom study and clinical correlation, Clin Radiol 68 (2013), e128–e135.

27.

McCollough

C.H.

, Yu

, Kofler

J.M.

, et al., Degradation of CT low-contrast spatial resolution due to the use of iterative reconstruction and reduced dose levels, Radiology 276 (2015), 499–506.

28.

Sookpeng

, Martin

C.J.

, Gentle

D.J.

Investigation of the influence of image reconstruction filter and scan parameters on operation of automatic tube current modulation systems for different CT scanners, Radiat Prot Dosimetry 163 (2015), 521–530.

29.

Dane

, Patel

, O’Donnell

, et al., Image quality on dual-energy CTPA virtual monoenergetic images: quantitative and qualitative assessment, Acad Radiol 25 (2018), 1075–1086.

30.

Hosseini Nasab

S.M.B.

, Shabestani-Monfared

, Deevband

M.R.

, et al., Estimation of cardiac CT angiography radiation dose toward the establishment of national diagnostic reference level for CCTA in Iran, Radiat Prot Dosimetry 174 (2017), 551–557.

31.

Lee

, Kim

H.J.

Noise properties of reconstructed images in a kilo-voltage on-board imaging system with iterative reconstruction techniques: A phantom study, Phys Medica 30 (2014), 365–373.

Effect of iterative reconstruction algorithm levels on noise index and figure-of-merit in CT pulmonary angiography examinations

Abstract

PURPOSE:

METHODS:

RESULTS:

CONCLUSION:

Keywords

1 Introduction

2 Materials and methods

2.1 Patient population

2.2 CT acquisition and protocols

Table 1 Data on scanning acquisition in CTPA examination Scanning Parameter Values Tube Voltage (Kv) 100 and 120 kV Tube Current (mAs)* 184.50±81.87 Scan Range (mm)* 277.44±88.03 Pitch Factor 0.798 Beam Collimation (mm) 0.625×40 Rotation Time (s) 0.50

3 Results

3.1 Radiation measurement and diagnostic performances

5 Conclusion

Footnotes

Acknowledgments

References

Table 1
Data on scanning acquisition in CTPA examination

Scanning Parameter Values

Tube Voltage (Kv) 100 and 120 kV

Tube Current (mAs)* 184.50±81.87

Scan Range (mm)* 277.44±88.03

Pitch Factor 0.798

Beam Collimation (mm) 0.625×40

Rotation Time (s) 0.50