Dose and Image Quality in C-arm
CT Rotational Angiography
R Fahrig, E Girard-Hughes , A Ganguly,
A Tognolini , N Kothary
Collaborators/Support
o The Physics Gang:: ‘Bob’ R.L.Dixon, ‘Tom’ J.
Payne, ‘Rick’ R.L. Morin
o A. Ganguly, N. Strobel and T. Moore
o L. Hoffman, N. Kothary, D. Sze, M. Marks, H. Do
o Technical support : A. White, N.R. Bennett, J.
Kneebone, M. Lozada-Parks, W. Baumgardner
o Siemens Medical Solutions
o NIH R01 EB003524
o Lucas Foundation
Department of Radiology, Stanford University
Introduction
o C-arm CT for visualization in 3D of vasculature
and other high-contrast structures has become
commonplace
o The transition from XRIIs to digital flat panels
opened the doors to the possibility of lowcontrast 3D CT imaging in the interventional
suite
o What doses are likely? How do current
settings compare to clinical CT doses?
o How is low-contrast visibility affected by
choice of kVp?
o How does the AEC system on the C-arm affect
dose and low-contrast visibility?
C-arm System :: CT System
half scan, area detector vs. full scan/narrow detector
Creating 3D Images in the
Interventional Lab
1) Rotational Angiography Run
4) In-room Display
Topics of Discussion
o First question : Dose vs. Image quality :
Neuroimaging
• Standard dosimetry with 16-cm head phantom
• image quality comparison as a function of dose
o Second question : Dose vs. Image quality :
Body Imaging
3) Reconstruction and
Visualization
2) Image transfer
Dose Measurement
o small 0.6cc Ion Chamber
o measuring maximum dose at center of z-extent of
the scanning range in an appropriate phantom
o See AAPM TG111 Report for full protocol
o Note that Farmer chamber calibration is typically
carried out using Co-60; special correction factors
are required for low-HVL dosimetry
• Automatic Exposure Control with custom phantom
• FOV reduction (slab imaging) improves conspicuity
o Third question : Does 3D information Increase
or Decrease dose in clinical use?
Farmer calibration
o non-standard calibration provided by Radcal
Corporation
kVp
70
81 109 125
HVL(mmAl) 2.9 3.2 3.4 4.4
Dose Measurement
o CTDI phantom (16cm diameter, 15cm long)
o Dose measured at center and eight peripheral
positions for :
(30x40) cm detector format
based on 543 views
o Beam Size (iso-center):
Width: 26.67cm
Height: 20.00cm
16 cm
NEUROIMAGING
32 cm
Dose Measurement
o CTDI phantom (16cm diameter, 15cm long)
o Dose measured at center and eight peripheral
positions for :
(30x40) cm detector format
based on 543 views
o Beam Size (iso-center):
Width: 26.67cm
Height: 20.00cm
16 cm
32 cm
Dose Measurements: 81kVp
2/3
mean
peripheral
+
1/3 central
Measured Doses :
‘Medium-High’ dose requested
Dose Measurements: 81kVp
D (0) = (1 / 3) D0 + ( 2 / 3) D p
2/3
mean
peripheral
+
1/3 central
source side
kVp
Total
mAs
Peak
Dose
(mGy)
86
Center
Dose
(mGy)
34
“CTDIw”
(mGy)
1167
Detector
dose
(uGy/view)
0.46
70
81
608
0.44
63
28
37
109
310
0.70
66
31
40
125
260
0.92
76
38
46
Variation in CTDIw in spite of AEC.
The EU guidelines for routine head CT scans specify a CTDIw
of 60mGy.
detector side
Visibility vs. Dose
Visibility of Low-Contrast Objects?
Visibility Chart (81kVp, 543view s)
Nominal Contrast
100
U)
(3H
90
80
9mm
0%
6mm
0
(1
)
HU
Detail Diameters [mm] (2, …, 9, 15)
70
Visibility (%)
15mm
0.5% (5HU)
%
0 .3
1.
o Catphan Module CTP515
used as image quality
phantom (20cm housing)
o Acquired 543 views over
20sec at various dose and
kVp settings, Zoom 0
o Reconstructed soft tissue
segment (smooth kernel,
10mm slice width)
o Analyzed visibility of (outer)
5HU insets
48
60
50
40
CTDIw = 19.93mGy
CTDIw = 26.31mGy
30
20
CTDIw = 37.01mGy
CTDIw = 54.12mGy
10
0
0
Scoring Question: What size “5HU” objects can you see?
3
6
9
Detail Diameter [mm]
12
15
d ref =
Normalized Visibility :
D ( 0)
⋅d
Dref (0)
Visibility vs. kVp
120
D (0)
⋅d
Dref (0)
120
100
Visibility [%]
100
80
Visibility [%]
d ref =
60
80
60
40
Average (70kVp)
40
Cat: ex 38.41mGy
Average (81kVp)
20
Cat: 56.63mGy
Average (109kVp)
Average (125kVp)
Cat: ex 81.73mGy
20
Average (56.63mGy)
0
0
2
4
6
8
10
12
14
16
Normalized Diameter [mm]
0
0
2
4
6
8
10
12
14
16
9mm scoring objects with contrast of
5HU visible in over 90% of all cases
Normalized Diameter [mm]
MTF: 100 µm steel wire
Intracranial Imaging
C-arm CT
Clinical CT
1.0
In vivo pig model,
Autologous blood,
NO iodine contrast
Sharp
0.8
Normal
MTF
Smooth
0.6
ARTIFACT:
0.4
Beam hardening
Scatter
Conebeam
0.2
0.0
0.0
0.5
1.0
1.5
2.0
2.5
Spatial Frequency (cycles/mm)
M. Marks, H. Do et al.
Conclusions - Neuro
Materials
Image Quality
o Place the ‘Siemens 16cm image quality ConeBeam’
phantom inside a body-shaped phantom
o 168 negative contrast objects in three slices :
3, 5, 10, 15, 20, 25, 30, 45, etc. HU,
32, 16, 8, 4 and 2 mm diameter
Image quality slices of
interest
BODY IMAGING
Body Phantom Design
50 cm
37 cm
Circumference
= 104cm
16 cm
26 cm
o Better visibility at lower energy (kVp) offers
potential for image acquisition protocol
optimization
o We can detect 9mm scoring objects (nominal
contrast “5HU”) in over 90% of all cases (70kVp
through 125kVp)
o The dose applied to obtain this image quality
performance is close to the EU guideline for CT
head scans (60mGy)
o The in-plane spatial resolution of the system is
excellent.
• IQ 16-cm conebeam phantom
was centered inside the body
phantom and two additional 50cm solid sections added to
provide adequate scatter
Dose Measurement Insert
• 13.1-mm diameter holes in 16-cm center
(same geometry as CTDI phantom)
• six holes at the half way points
• the eight holes at the edge of the green
outer ring are
placed such that
distance from
center of the
hole to the edge
of phantom = 10 mm
AEC kVp for 109 kVp Request
Experiment
o Imaged at 3 different collimations :
28 cm (22.2 cm @ isocenter)
17.5 cm (14 cm @ isocenter)
10.5 cm (8.4 cm @ isocenter)
o Smallest collimation still included the full
dominant for automatic exposure control
o DynaCT, 0.5 degrees per step, large focus,
8-s scan, 396 projections/reconstruction
o kVp = 109 or kVp = 125 kVp
o dose request = 0.36 uGy, 0.54 uGy, or
0.81 uGy per image at the detector
AEC mAs
Impact of Scatter on Image
Quality
o Images of a 16-cm ‘medium contrast’
insert in the abdomen phantom, 0.8 mm
slice thickness
o Objects are : -20, -25, -30 and -45 HU
32, 16, 8, 4 and 2 mm diam
109 kVp
125 kVp
Impact of Scatter on Image
Quality
o Images of a 16-cm ‘low contrast’ insert in
the abdomen phantom, reconstructed
with 5.0 mm slice thickness
o Objects are : -3, -5, -10 HU and -15 HU
32, 16, 8, 4 and 2 mm diam
Detectibility @ 125 kVp
o Sum up the number of visible objects in each of
the three slices : high-contrast, medium-contrast
and low-contrast : recon 0.5x0.5x1.0 mm
o Three ‘group’ reads of each dataset at 125 kVp
as a fn. of collimation
as a fn. of dose requested
Conclusions – Body Imaging
o Need to calibrate AEC behavior for a
range of body sizes; our phantom
represents a very un-circular (slim?) case
o Body imaging pushes the x-ray tube
limits :
• Higher kVp needed to increase tube output
per mAs and to increase patient penetration
CLINICAL DOSES
o Slab imaging increases detectibility and
decreases average dose in max. slice (as
well as decreasing total vol. irradiated)
TACE Rationale
Chemoembolization:
Procedure/Technique
o Ischemia to tumor
• Susceptible to therapy
o Increase
concentration
compared to IV
infusion
o Increased dwell time
o Reduced systemic
toxicity
o Aortogram
o Visceral arteriography
• Delineate vessels (i.e.
cystic artery)
o Selective catheterization
of target artery
supplying tumor
L. Hoffman et al.
Augmented by 3D
Chemoembolization:
A clinical study
o Acquisition: 419 images, 8 second scan through 210°°,
512x512 reconstruction, 0.36 µGy per frame
o Injection : 50-50 mix of iodinated contrast and saline
o technically successful in 93/100 procedures (93%)
o provided information not available by DSA in 30
patients (35.7%)
o resulted in a change in diagnosis, treatment planning
or treatment delivery in 24 patients (28.6%)
L. Hoffman et al.
Methods- Imaging protocol
Methods- Imaging protocol
DSA study arm (reference standard):
C-arm CT study arm:
o DSA groin run at 42 cm FOV
Aortogram 30cc @ 15 cc/sec
o Non-contrast CACT if patient had residual ethiodol
from previous chemoembo
o AP and 30º RAO, 21 ml @ 3ml/sec from
proper/common hepatic artery (~ 3 frame/s, w/
late parenchymal phase) POWER INJECTION
ONLY.
o Proper hepatic CACT 12cc from proper/common
hepatic (4 sec X-ray delay)- 8 sec rotational scan (~
200°) with image acquisition every 0.5° (419 images)
o Additional DSA as required
o Additional oblique DSA as needed
o Single DA image of the RUQ and completion
post-embolization angiogram (hand)
o Additional selective CACT if needed for problem
solving
o Post-treatment non-contrast CACT
Methods: C-arm CT
Navigation
Methods- data collection
Methods- additional data collection
Radiation dose endpoints (dose report):
o Total Dose Area Product (DAP: µGym²)
o DAP for guidance
• from 5F catheter in common/proper hepatic to
superselective positioning of microcatheter
o Fluoro time (min) for total procedure
o Fluoro time for guidance
Time endpoints (procedure log sheet):
o Total procedure time (min)
o Time for guidance
o For DSA arm
• Contrast enhanced CACT performed,
however images not reconstructed till
microcathter positoned
• Post chemoembolization non-contrast CACT
(instead of unenhanced helical CT) to ensure
complete geographic uptake in the tumor
Methods- DSA study arm
Methods- CACT study arm
Total DAP: 14965.6 µGym²
Guidance DAP: 4425.6+860.3= 5285.9 µGym²
Total DAP: 35812.2– (6559.6+6499.3)= 22753.3 µGym²
Guidance DAP: 3944.6+6260.9+1967.6= 12173.1µGym²
Results - DAP (µGym²)
CACT study arm
Mean: 16513.9
Guidance Median:14690.3
(range: 5012-33628)
Total
dose
(includes
fluoro )
DSA study arm
Mean: 15143.5
9.5%
Median:13406.6
(p= 0.588)
(range: 2567-30684)
Mean: 36567.4
Mean: 27417. 2
Median: 31945.1
Median: 25134.1
(range:14716-65270) (range:4717-65786)
(*includes w/ CACT
post)
% dose
increment
(median)
27%
(p= 0.069)
(mean DAP w/CACT 11%
post: 32917.2)
(p=0.4)
Results - Total and Guidance Time
CACT study arm
DSA study arm
% increase (median)
Total time
(min)
Mean: 100
Median:90 (range:62171)
Mean: 80
Median:76
(range: 30-141)
18.4%
(p= 0.08)
Total time
from CHA to
completion
(min)
Mean: 65
Median:58
(range: 32- 108)
Mean: 57
Median:54
(range: 27-98)
7.4%
(p= 0.19)
Guidance
time (min)
Mean: 33
Median: 26
(range: 7-51)
Mean: 31
Median: 25
(range: 9-74)
4%
(p= 0.54)
Results : Fluoro time/Contrast volume
CACT study arm
DSA study arm
% increase
(median)
Total fluoro
time
(min)
Mean: 26
Median: 23
(range: 9-65)
Mean: 22
Median: 20
(range:10-43)
15%
(p= 0.17)
Fluoro time
guidance
(min)
Mean: 8
Median: 6
(range: 2-20)
Mean: 20
Median:17
(range: 9-40)
-64%
(p= 0.0001)
Contrast- P.
Injector
(ml)
Mean: 48
Median: 45
(range:27-75)
Mean: 86
Median: 82
(range: 63-120)
-45%
(p= 0.0001)
Contrasttotal
(includes
hand runs)
Mean: 78
Median: 75
(range:72-120)
Mean: 95
Median: 100
(range: 60-130)
-25%
(p= 0.0001)
Conclusion
o C-arm CT does not significantly add to
the radiation dose, procedural time or
catheterization time
o C-arm CT potentially can reduce operator
dose by decreasing fluoroscopy time
o C-arm CT reduces overall contrast
volume
• Particularly important in this group of
patients that are predisposed to
nephrotoxicity due to underlying cirrhosis
Context
CARDIAC IMAGING
o Single sweep in the interventional suite
provides geometry of cardiac chambers
for guidance of RF ablation
o Contrast injected
into the chamber of
interest during acq.
o System lateral
• tube rotates from +10º
above horizontal,
around patient back to
+10º above horizontal
Automatic Registration
o pre-registration of EAM system and C-arm CT FOV
could shorten RF procedures by eliminating timeconsuming point-by-point registration as is done
currently with prior CT or MR images
Monte Carlo-based Dose
o Excellent paper :
“Effective dose analysis of three-dimensional
rotational angiography during catheter ablation
procedures”
• J-Y Wielandts, K Smans, J Ector, S De Buck, H Heidbuchel and H
Bosmans
Phys. Med. Biol. 55 (2010) 563–579
o Monte Carlo used to calculate BMI-specific
organ doses using recorded (and variable!)
kVp/mAs and assumed geometry
o 5-s sweep with 249 projections, 60 fps
Monte-Carlo Results
Multi-sweep Gated…
From Phys. Med. Biol. 55 (2010) 563–579
o Specific organ doses that are of interest
due to highest weights
• Breasts, lung, bone marrow (sternum), lung
o Timing the return of each rotation properly provides
sufficient data for a reconstruction of ¼ of the cardiac
cycle e.g. in diastole
4x5s Sweep, ECG gating
Large (30x40 cm) Detector
Patient 6:
with ECG gating
o 4 sweeps, 5s per sweep,
~250 projections per
sweep
o Total scan time ~24 s
(including time for C-arm
turn around) : total
breath hold ~ 27 s
o 12-15 s delay between
start of injection and
start of imaging
o Injection of ~140 ml
contrast (3:1 dilution) at
4 ml/s into inferior vena
cava
Patient 6:
no ECG gating
Comparison of Protocols
Rando Phantom Organ Dose?
Multi-sweep Cardiac Dose?
o Measured organ doses according to standard
prescriptions for dosimeter locations (solid state
MOSFET dosimetry system calibrated for diagnostic
energies) but for our imaging protocol
o Compared against
Organ thyroid esophagus lung heart
MC simulations
seems to be
Rando
0.44
5.71
5.93 6.80
significantly lower
(mGy)
Monte
o Based on 4xMCavg
Carlo
0.94
13.05
23.35 16.6
for 82 kg person
(mGy)
our gated cardiac
scans would give
Scale
0.47
0.44
0.25 0.41
22 mSv
o Based on 4xMCavg for 82 kg person our
gated cardiac scans would give 22 mSv
o Clearly individual organ scaling factors
indicate the dose will be lower
o Clinical un-gated CT cardiac scans are
between 10-16 mSv
o Our ‘constant energy’ system setting was
0.18 kJ vs. MC-paper value of 0.25 kJ
o Need to expand Rando measurements to
‘large’ patients to stress the system
From Phys. Med. Biol. 55 (2010) 563–579
C-arm CT vs. Clinical CT
Image quality
• Streak artifacts due to remaining motion
• Contrast brightening and darkening due to beam
hardening and scatter
o
o
o
o
o
o
o
Intra-procedural
Single circle scan
1000 slices
Rotation time 5-10s
Volume in 5-10s
See better than 10 HU
AEC!
o
o
o
o
o
o
o
Diagnostic
Spiral scan
64 slices
Rotation time 0.3s
Volume in 5-10s
See better than 3 HU
mA modulation?
Conclusions
o Dosimetry for rotation C-arm CT
angiography is complicated by AEC
o Careful evaluation of dose vs. image
quality for a particular application is
required
o We have the necessary tools…