Download CT chest and gantry rotation time

Original Article
CT chest and gantry rotation time: does
the rotation time influence image quality?
Acta Radiologica
2015, Vol. 56(8) 950–954
! The Foundation Acta Radiologica
2014
Reprints and permissions:
sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/0284185114544242
acr.sagepub.com
Martin Beeres1,*, Julian L Wichmann1,*, Jijo Paul1,
Emmanuel Mbalisike1, Mohamed Elsabaie1, Thomas J Vogl1 and
Nour-Eldin A Nour-Eldin1,2
Abstract
Background: Computed tomography (CT) gantry rotation time is one factor influencing image quality. Until now, there
has been no report investigating the influence of gantry rotation time on chest CT image quality.
Purpose: To investigate the influence of faster gantry rotation time on image quality and subjective and objective image
parameters in chest CT imaging.
Material and Methods: Chest CT scans from 160 patients were examined in this study. All scans were performed
using a single-source mode (collimation, 128 0.6 mm; pitch, 1.2) on a dual-source CT scanner. Only gantry rotation
time was modified, while other CT parameters were kept stable for each scan (120 kV/110 reference mAs). Patients
were divided into four groups based on rotation time: group 1, 1 s/ rotation (rot); group 2, 0.5 s/rot; group 3, 0.33 s/rot;
group 4, 0.28 s/rot. Two blinded radiologists subjectively compared CT image quality, noise, and artifacts, as well as
radiation exposure, from all groups. For objective comparison, all image datasets were analyzed by a radiologist with
5 years of experience concerning objective measurements as well as signal-to-noise ratio (SNR).
Results: We found that faster gantry rotation times (0.28 s/rot and 0.33 s/rot) resulted in more streak artifacts, image
noise, and decreased image quality. However, there was no significant difference in radiation exposure between faster
and slower rotation times (P > 0.7).
Conclusion: Faster CT gantry rotation reduces scan time and motion artifacts. However, accelerating rotation time
increases image noise and streak artifacts. Therefore, a slower CT gantry rotation speed is still recommended for higher
image quality in some cases.
Keywords
Chest computed tomography (CT), gantry rotation time, CT artifacts, CT image quality, CT radiation exposure
Date received: 6 December 2013; accepted: 21 June 2014
Introduction
Computed tomography (CT) imaging of the chest is
one of the most frequent radiologic examinations in
the world, as it is a non-invasive way in which to evaluate a variety of clinical issues, such as pulmonary infiltration or potential malignancies. Even though multiple
techniques are designed to reduce radiation exposure,
chest CT imaging still exposes patients to a higher
amount of radiation than other imaging modalities
(e.g. X-ray). However, radiation exposure in chest CT
imaging has decreased remarkably in recent years using
low kV imaging, iterative reconstruction algorithms,
and campaigns by radiological societies around the
world, such as the ‘‘Image Wisely’’ campaign from
the Radiological Society of North America and
American College of Radiology (1–3).
1
Department of Diagnostic and Interventional Radiology, Clinic of the
Goethe University, Frankfurt, Germany
2
Diagnostic and Interventional Radiology Department, Cairo University
Hospital, Cairo, Egypt
*These authors are equal contributors.
Corresponding author:
Martin Beeres, Department of Diagnostic and Interventional Radiology,
Clinic of the Goethe University, Haus 23C UG, Theodor-Stern-Kai 7,
60590 Frankfurt, Germany.
Email: [email protected]
Downloaded from acr.sagepub.com at Universitaetsbibliothek on July 18, 2016
Beeres et al.
951
State-of-the-art CT scanners implement rapid imaging strategies to reduce radiation exposure, including
wide detectors, fast gantry rotation times, two detectors, or a combination of these strategies (3–5). In
nearly every CT machine, fast tube detector rotation
times of 0.5 s / rotation (rot) are possible and can be
used in different clinical situations. Fast gantry rotation
speeds combined with an adapted pitch-factor shorten
image acquisition times, which is advantageous when
imaging the thoracic region because it reduces the
number of motion artifacts. Unfortunately, accelerating the gantry rotation time can also introduce additional artifacts (e.g. streaking) that are not visible
using slower speeds. Therefore, the aim of this study
was to investigate the influence of gantry rotation time
on image quality, noise, artifacts, and radiation exposure during chest CT examinations.
Material and Methods
Patients and CT protocols
110 reference mAs. Gantry rotation times for each
group were as follows: group 1, 1 s/rot; group 2, 0.5 s/
rot; group 3, 0.33 s/rot; and group 4, 0.28 s/rot.
Automatic exposure control was used for all scans
(CARE Dose 4D, Siemens Healthcare, Forchheim,
Germany). Data were acquired in the caudocranial
direction during deep inspiratory breath-hold.
Transverse images were reconstructed with a 5 mm
slice thickness in 5 mm increments using a hard convolution kernel in filtered back-projection (B60f), a matrix
of 512 512, and a lung window (center, –500 HU;
width, 2000 HU) for the overview. For further analysis,
transverse 1.0 mm slices in 0.7 mm increments were
reconstructed. For 3D evaluation, coronal and sagittal
reformations with a 2 mm slice thickness in 2 mm increments were reconstructed. Additional patient parameters such as body mass index (BMI) and scan range
were also evaluated.
Image analysis
This single-center, observer-blinded, retrospective study
was approved by the local Ethics Committee and written informed consent was obtained by all patients. Data
of consecutive patients that underwent clinically indicated CT of the chest between November 2011 and
April 2013 were analyzed. All CT examinations
included in this study were diagnostic and did not
require repeating because of unsatisfactory image quality or artifacts.
Patients were divided into four groups of 40 individuals each based on gantry rotation time (Table 1). All
groups underwent chest CT scanning on a dual-source
CT operated in single-source mode functioning as a
128 slice CT (Somatom Definition Flash, Siemens
Healthcare, Forchheim, Germany) with a 1.2 pitch,
128 0.6 mm collimation, 120 kV tube potential, and
The total examination time of each CT image series was
recorded in seconds. Objective image quality was
assessed using several region-of-interest (ROI) measurements (1 cm2) as standard deviation (SD) at different anatomic levels (lung apex, pulmonary trunk,
phrenicocostal sinus) taken by a radiologist with
5 years of CT experience on a regular PACS (Picture
Archiving and Communication System) workstation
(Centricity 4.2, GE-Healthcare, Dornstadt, Germany)
using a circle tool. Subjective image evaluation was
carried out by two independent radiologists (with
5 years and 4 years of CT imaging experience) regarding motion and streak artifacts, pathological visibility,
and image quality. Image quality rating was performed
using a five-point Likert scale (5, excellent; 4, good; 3,
moderate; 2, fair; 1, unacceptable). Additional ROI
evaluations were performed in the subcutaneous fat at
Table 1. Examination parameters.
Slice collimation
Pitch
kV / ref. mAs
Rotation time
(seconds/rotation)
Patients
Male
Female
Age (years)
BMI (kg/m2)
Group 1
Group 2
Group 3
Group 4
128 0.6
1.2
120 / 110
1.0
128 0.6
1.2
120 / 110
0.5
128 0.6
1.2
120 / 110
0.33
128 0.6
1.2
120 / 110
0.28
40
24
16
67 (25–85)
26.3 3.4 (18.7–31.3)
40
23
17
63.5 (26–87)
27.3 3.0 (19.5–32.5)
40
20
20
70 (36–90)
28.1 2.9 (20.2–31.1)
40
27
13
63 (42–77)
27.2 3.1 (19.1–30.8)
Downloaded from acr.sagepub.com at Universitaetsbibliothek on July 18, 2016
952
Acta Radiologica 56(8)
the level of the pulmonary artery and muscle tissue;
Musculus erector spinae muscles were chosen. To minimize bias from single measurements, we calculated
the average of four measurements. Based on
these measurements, signal-to-noise ratio (SNR) was
determined according to the following equation:
SNR ¼ Attenuation/Background Noise.
radiation dose. The Mann-Whitney U test was calculated using Bonferroni correction.
A P value less than 0.05 was considered statistically
significant. A Cohen’s kappa (weighted kappa) analysis
was performed to determine inter-observer agreement
for subjective image quality scoring.
Results
Radiation exposure
To estimate patient radiation dose, we recorded the
volume CT dose index (CTDIvol, mGy) and dose
length product (DLP, mGy*cm) from the patient CT
protocol.
Statistical analysis
All statistical analyses were performed using BiAS 9.14
software (BiAS 9.14, epsilon-Publishing, Hochheim,
Germany). Continuous variables were expressed as
median and range; categorical variables were expressed
as frequencies or percentages. The Kruskal-Wallis test
and for post-hoc testing the Mann-Whitney U test was
used to compare image noise, attenuation, and
The median exam duration in group 4 (2.4 s; range,
0.6 s) was significantly shorter compared to groups 1
(7.7 s; range, 1.6 s; P < 0.05) and 2 (4.2 s; range, 1.5 s;
P < 0.05; Table 2). In addition, the median exam duration was significantly shorter in group 1 relative to
group 3 (2.6 s; range, 0.6 s; P < 0.05; Table 2).
Furthermore, there was no statistical difference in
BMI (P > 0.5; Table 1), DLP, CTDIvol, or mAs
(P > 0.05; Table 2) between groups.
Evaluation of image noise as an objective image
quality parameter showed that slower gantry rotation
time protocols in groups 1 and 2 resulted in significantly less image noise compared with faster rotation
times in groups 3 and 4 (P < 0.01; Table 2; Fig. 1). The
SNR-values were influenced additionally to the image
Table 2. Examination parameters.
Group 1
Group 2
Group 3
Group 4
Scan range (cm)
Scan duration (s)
35.5 (31.9–39.3)
7.7 (6.9–8.5)
38.5 (30.8–44.5)
4.2 (3.3–4.8)
36.5 (32.0–40.6)
2.6 (2.3–3.1)
38.8 (31.7–45.3)
2.4 (2.2–2.9)
Dose-length product
(mGy cm)
CTDI vol.
mAs
Image noise (HU)
271 (171–443)
266 (143–323)
253 (166–483)
236 (176–311)
7.4 (4.6–17.8)
110 (68–265)
24.8 (16.0–30.8)
7.2 (4.2–13.5)
110 (69–205)
27.0 (23.1–35.8)
7.5 (4.7–13.9)
105 (62–200)
34.3 (33.1–35.9)
6.8 (3.8–10.4)
99 (55–130)
35.8 (32.6–47.8)
2.0 (1.4–3.1)
1.8 (1.2–2.8)
1.4 (1.3–1.6)
1.5 (0.6–1.6)
SNR
Downloaded from acr.sagepub.com at Universitaetsbibliothek on July 18, 2016
P value: Group 1
vs. Group 2/3/4
0.06
1 vs.
1 vs.
1 vs.
2 vs.
2 vs.
3 vs.
0.1
0.6
0.4
1 vs.
1 vs.
1 vs.
2 vs.
2 vs.
3 vs.
1 vs.
1 vs.
1 vs.
2 vs.
2 vs.
3 vs.
2:
3:
4:
3:
4:
4:
0.3;
0.003;
0.0005
0.1;
0.05;
0.6
2:
3:
4:
3:
4:
4:
2:
3:
4:
3:
4:
4:
0.2;
<0.01;
<0.01
<0.01;
<0.01;
0.3
0.05;
<0.01;
<0.01
<0.01;
0.02;
0.5
Beeres et al.
953
Fig. 1. (a) 0.28 s / rot. rotation time – artifacts are present; (b) s / rot. rotation time – artifacts still present; (c) 0.5 s / rot. rotation
time – less noise and artifacts; (d) 1.0 s / rot. rotation time – less artifacts compared to (a) and (b).
Table 3. Image quality rating between the different groups.
Streak artefacts
Kappa
Overall quality
Kappa
Observer 1
Observer 2
Observer 1
Observer 2
Group 1
Group 2
Group 3
Group 4
1 0.5
1 0.5
0.93
2 0.6
1 0.5
0.40
1 1 (1–3)
2 0.5 (1–3)
0.82
1 0.7 (1–3)
2 0.6 (1–3)
0.64
2 0.5
2 0.8
0.78
2 0.6
2 0.7
0.75
2 0.9 (1–3)
3 1 (1–4)
0.65
2 0.8 (1–3)
2 0.7 (1–3)
0.8
(1–2)
(1–3)
(1–3)
(1–2)
(1–3)
(1–3)
(1–3)
(1–3)
Overall
Kappa
0.8
0.6
Fig. 2. Sufficient image quality in all cases: Four patients: (a) 1.0 s / rot. rotation time – nearly no noise artifacts in the picture and
surroundings; (b) 0.5 s / rot. rotation time – less noise artifacts in the picture and surroundings; (c) 0.33 s / rot. rotation time –
increasing noise artifacts in the picture and surroundings; (d) 0.28 s / rot. rotation time – most artifacts in the picture and surroundings.
noise (Table 2). The best image quality was obtained
with 0.5 and 1 s/rot rotation times (Table 3), although
image quality differences were not significant overall
(Fig. 2). Interestingly, only Radiologist/Observer 2
found significant differences in subjective image analysis ratings between groups 1 and 3 (P ¼ 0.007) and
groups 1 and 4 (P ¼ 0.04; Table 3), the rest did not
reach statistical significance. However, there was an
obvious increase of image noise and streak artifacts in
higher gantry rotation time groups (Tables 2 and 3). As
image noise increased, artifacts became more visible,
which influenced subjective image scoring in some
cases.
Discussion
Ongoing research has shown that increasing detector
width and rotation speed not only make CT imaging
faster, but also improves workflow, making this
imaging modality more robust against motion artifacts
(4,6). Motion is a very common and significant problem
in CT imaging because patients are not always able to
hold their breath for a defined period of time or may
simply be non-compliant (4). Unfortunately, accelerating rotation speed can introduce additional artifacts,
noise, or other image quality issues that are not visible
at slower speeds. Therefore, we investigated the influence of gantry rotation time on various imaging parameters and radiation exposure during chest CT
examinations.
Our study shows for the first time that the slowest
tube rotation time setting (1 s/rot) results in a breathhold period of 6–9 s (Table 2). All patients examined in
this mode were able to and did hold their breath for
this period of time. Although most previously published reports focus on ‘‘non-breath-hold’’ imaging of
the chest (7,8), data on both ‘‘breath-hold’’ and ‘‘nonbreath-hold’’ situations is important for adaptation of
Downloaded from acr.sagepub.com at Universitaetsbibliothek on July 18, 2016
954
Acta Radiologica 56(8)
CT protocols in different clinical situations. The SNR
values showed some differences, so group 1 and
2 offered the highest values compared to group 3 and
4 (Table 2).
A previous abdominal imaging report from 2008
demonstrated that using a faster tube-detector rotation
time saturates the X-ray tube more often (at its limit of
current), leading to an increase in image noise and
deterioration of image quality (9). Conversely, a 2012
report concluded that rotation time and pitch factors
have only a small influence on image quality (10).
Present results agree with both studies in that
although faster tube-detector rotation time minimally
influenced overall image quality, image noise significantly increased. Although image quality was highest
at longer gantry rotation times (0.5 and 1 s/rot;
Table 3), differences between rotation times were not
significant (Fig. 2). On the other hand, image noise was
significantly higher using faster rotation (Tables 2 and
3). Moreover, with increasing image noise, artifacts
sometimes become more visible, which influences subjective image scoring. Furthermore, image noise might
influence image quality differently according to the
body region examined.
Although a shorter CT scan time would technically
reduce the radiation exposure time, it is unclear
whether faster gantry rotation directly affects the
absorbed dose. Interestingly, a 2006 report posited
that using faster gantry rotation times at a high tube
load might lead to a larger beam focus, which might
increase overall radiation exposure (11). However, our
data demonstrated that there was no statistical difference in DLP or CTDIvol between rotational times
investigated in this study (P > 0.05; Table 2).
In conclusion, faster CT gantry rotation speed
reduces scan time and motion artifacts. However, current results, together with previous studies, demonstrate that accelerating rotation time does not always
provide the best image quality, noise, or artifact presentation. Furthermore, in CT machines with wide
detectors, a fast gantry rotation is not always necessary
to acquire images in a short period of time. Therefore, a
slower CT gantry rotation speed is still recommended
for higher image quality in some clinical cases.
Funding
This research received no specific grant from any funding
agency in the public, commercial, or not-for-profit sectors.
References
1. Brink JA, Amis ES Jr. Image Wisely: a campaign to
increase awareness about adult radiation protection.
Radiology 2010;257:601–602.
2. Dougeni E, Faulkner K, Panayiotakis G. A review of
patient dose and optimisation methods in adult and
paediatric CT scanning. Eur J Radiol 2012;81:665–683.
3. Fleischmann D, Boas FE. Computed tomography–old
ideas and new technology. Eur Radiol 2011;21:510–517.
4. Kroft LJ, Roelofs JJ, Geleijns J. Scan time and patient
dose for thoracic imaging in neonates and small children
using axial volumetric 320-detector row CT compared to
helical 64-, 32-, and 16- detector row CT acquisitions.
Pediatr Radiol 2010;40:294–300.
5. Lell MM, May M, Deak P, et al. High-pitch spiral computed tomography: effect on image quality and radiation
dose in pediatric chest computed tomography. Invest
Radiol 2011;46:116–123.
6. Hinkmann FM, Voit HL, Anders K, et al. Ultra-fast
carotid CT-angiography: low versus standard volume
contrast material protocol for a 128-slice CT-system.
Invest Radiol 2009;44:257–264.
7. Renne J, Falck CV, Ringe KI, et al. CT angiography for
pulmonary embolism detection: the effect of breathing on
pulmonary artery enhancement using a 64-row detector
system. Acta Radiol. Epub ahead of print 8 October
2013. DOI: 10.1177/0284185113507712.
8. Schulz B, Jacobi V, Beeres M, et al. Quantitative analysis
of motion artifacts in high-pitch dual-source computed
tomography of the thorax. J Thorac Imaging 2012;27:
382–386.
9. Israel GM, Herlihy S, Rubinowitz AN, et al. Does a
combination of dose modulation with fast gantry rotation time limit CT image quality? Am J Roentgenol 2008;
191:140–144.
10. Klink T, Regier M, van Stevendaal U, et al. Accelerating
image acquisition in 64-MDCT: the influence of scan parameters on image resolution and quality in a phantom
study. Clin Imaging 2012;36:334–344.
11. Theocharopoulos N, Perisinakis K, Damilakis J, et al.
Dosimetric characteristics of a 16-slice computed tomography scanner. Eur Radiol 2006;16:2575–2585.
Conflict of interest
None declared.
Downloaded from acr.sagepub.com at Universitaetsbibliothek on July 18, 2016