Performance evaluation for new-technology CT systems
256-Channel CT
Performance evaluation for new-technology CT systems
256-Channel CT
Koos Geleijns
Leiden University Medical Center
Leiden, The Netherlands
1x5
4x1
16 x 0.5
64 x 0.5
256 x 0.5
1998
2001
2004
Soon ?
Performance evaluation for new-technology CT systems
Performance evaluation for new-technology CT systems
256-Channel CT
256-Channel CT
Almost 130 mm
• Medical physicists are challenged to
assess the new technology of 256Channel CT scanners:
– Current concepts for measurement of image
quality still remain valid (PSF, MTF, NPS)
256 x 0.5
– Concepts for CT dosimetry will have to
change fundamentally
Soon ?
Heart
1
Performance evaluation for new-technology
CT systems 256-Channel CT
• First: experimental 256 Channel CT Scanner at the
Japanese National Institute of Radiological Sciences
• Now also: Toshiba prototype 256 Channel CT scanner
Toshiba prototype 256 Channel CT scanner
Methods for dosimetry with 256-Channel CT
• Characterisation of the x-ray beam
Two wedge filters: 320 mm FOV and 500 mm FOV
Medium Field Of View (320 mm)
– Measurements free in air
• Dosimetry in phantoms
– Cilindrical CT dose phantoms
• Patient dose
– Monte Carlo dosimetry and voxel phantoms
Large Field Of View (500 mm)
2
Characterisation of the x-ray beam
• Dosimetry free in
air
– Small ionisation
chamber (thimble,
2.4 cm lenght)
– Film dosimetry
(GafChromic XR-QA)
– Extended CT
ionization chamber
(extended pencil, 30
cm lenght)
– Optically stimulated
luminescence (OSL)
dosemeter
X-ray tube: fixed
Thimble chamber
Thimble chamber
Z-axis
GafChromic
300 mm
extended pencil
chamber
XR-QA
25.4 x 30.5 cm
X-axis
OSL
X-axis (axial plane) profile in air
Z-axis profile in air
1.2
120 kV - Large FOV
80 kV - Large FOV
120 kV - Medium FOV
80 kV - Medium FOV
120 kV - L mirrorred
80 kV - L mirrorred
120 kV - M mirrorred
80 kV - M mirrorred
120 kV - Large FOV
80 kV - Large FOV
80 kV - Medium FOV
120 kV - Medium FOV
1.0
Thimble chamber
1
0.8
Normalised dose, unit less
Dose, unitless
0.8
0.6
0.4
0.6
0.4
Anode
Kathode
0.2
0.2
0.0
-20
0
Thimble chamber
-220
-170
-120
-70
30
x-axis, mm
80
130
180
-100
-80
-60
-40
-20
0
20
40
60
80
100
z-axis, mm
3
X-ray tube: fixed
Film dosimetry
GafChromic
XR-QA
XZ-axis plane
25.4 x 30.5 cm
• Gafchromic XR-QA film
• Sensitive in the 0 – 100 mGy range
• 2D dose profile
• X-ray tube: fixed at 12 o’clock position
GafChromic XR-QA dosimetry and
calibration
GafChromic XR-QA film calibration
curve
120
Entire FOV, fixed x-ray tube, film at the level of the Center Of Rotation (COR)
0 mGy
11.3 mGy
22.5 mGy
45.1 mGy
70.5 mGy
90.2 mGy
Absorbed dose, mGy
100
120 kV
80
80 kV
60
40
20
0
0
5
10
15
20
25
30
Color, red channel
4
GafChromic XR-QA 2D dose
distribution
Comparison of thimble chamber and Gafchromic
XR-QA film: profile along the x-axis
16
80 kV - Large FOV - GafChromic
Absorbed dose,
mGy
14
80 kV - Large FOV - Thimble Chamber
12
10
8
6
4
2
0
-240 -200 -160 -120
-80
-40
0
40
80
120
160
200
240
Position along x-axis, mm
Comparison of thimble chamber and Gafchromic
XR-QA film: profile along the z-axis
X-ray tube: fixed
18
Absorbed
dose, m Gy
16
Z-axis
14
80 kV - Large FOV - GafChromic
12
80 kV - Large FOV - Thimble Chamber
10
8
6
4
2
0
-80
-60
-40
-20
0
20
40
60
80
Position along the z-axis, mm
5
OSL dosimetry free-in-air
120 kV medium FOV
(courtesy Mike McNittGray and Rich Mather)
1.2
OSL
Thimble chamber
Relative absorbed dose, unitless
1
0.8
0.6
0.4
0.2
-100
-50
0
0
Position along the z-axis, mm
50
100
X-ray tube: rotating
300 mm CT pencil chamber
300 mm
extended pencil
chamber
6
Dosimetry in phantoms
The “correct” way:
• Extended (350 mm) CT dose phantoms: head and
body
• 64 Channels
• CT Head Phantom
– Extended CT ionization chamber (extended pencil, 300 mm
lenght); or
– Optically stimulated luminescence (OSL) dosemeter
• 150 mm Long
Phantom
The “incorrect” way:
• Regular (150 mm) CT dose phantoms: head & body
• 100 mm Pencil
Chamber
– Regular CT ionization chamber (pencil, 100 mm length)
The “correct” way
• 256 Channels
• 256 Channels
• CT Head Phantom
• CT Head Phantom
• 150 mm Long
Phantom
• 350 mm Long
Phantom
• 100 mm Pencil
Chamber is too small
….
• 300 mm Pencil
Chamber
7
• 64 Channels
• 256 Channels
• CT Body Phantom
• CT Body Phantom
• 150 mm Long
Phantom
• 150 mm Long
Phantom is too small
….
• 100 mm Pencil
Chamber
• 100 mm Pencil
Chamber is too small
….
The “correct” way
• 256 Channels
• CT Body Phantom
• 350 mm Long
Phantom
• 300 mm Pencil
Chamber
8
CTDI300: 300 mm pencil chamber
(courtesy Yoshihisa Muramatsu)
• CTDI in two phantoms
The “incorrect” way in a head phantom
NOT CTDI but an average dose
• 256 Channels
• CT Head Phantom
• 150 mm Long
Phantom
• 100 mm Pencil
Chamber
D × 300mm
CTDI =
256 × 0.5mm
Average dose (D) in a 150 mm long head
phantom measured with a 100 mm CT chamber
relative to CTDI300 (CTDI)
The “incorrect” way in a body phantom
NOT CTDI but an average dose
• 256 Channels
kV
Wedge Phantom Center Periphery Weighted
D/CTDIc D/CTDIp D/CTDIw
• CT Body Phantom
80
L
head
82%
93%
89%
80
M
head
83%
92%
89%
• 150 mm Long
Phantom
120
L
head
82%
93%
89%
• 100 mm Pencil
Chamber
120
M
head
82%
94%
90%
9
Average dose (D) in a 150 mm long body
phantom measured with a 100 mm CT chamber
relative to CTDI300 (CTDI)
kV
Wedge Phantom Center Periphery Weighted
D/CTDI D/CTDI
D/CTDIw
80
L
body
69
104%
97%
80
M
body
69
103%
96%
120
L
body
67
99%
91%
120
M
body
67
99%
91%
OSL Landauer (w.i.p.)
(courtesy Mike McNittGray and Rich Mather)
Monte Carlo dosimetry
courtesy Marcal Salvado
• OSL dosimeters
• Z-axis dose profile (in air and in phantoms)
10
Monte Carlo dosimetry
Monte Carlo dosimetry
Using EGS4 code
(Electron Gamma Shower)
Checking of MC dosimetry against dose
measurents (thimble chamber): axial plane
CT Scanner model:
wedges, collimators,
inherent filtration … and
heel effect
Monte Carlo dosimetry
Monte Carlo dosimetry
Checking of MC dosimetry against dose
measurents (thimble chamber): z-axis profile
CT Cardio and CT Brain examinations in voxel phantoms
Segmented in organs for calculation of effective dose
40
Measurements
Simulation without Heel effect
35
5 Materials:
- Air
- CT table
- Lung
- Bone
- Muscle
Simulation with Heel effect
Dose (mGy/100 mAs)
30
25
20
15
10
5
Small Female
0
-10
-8
-6
-4
-2
0
2
4
6
8
10
Position (cm)
11
Conclusion
• Dosimetry for 256 channel CT scanners
requires new concepts
• Problem: the large cone beam becomes
larger than the CT dose (body) phantom
and the 100 mm CT pencil ionization
chamber
• Advantage: the large beam allows for
measurements with small (thimble)
ionization chambers
Characterisation of the x-ray beam
• Dosimetry free in air
– Small ionisation
chamber (thimble, 24
mm lenght)
– 2D dosimetry
(GafChromic XR-QA,
but also CR)
– Extended CT
ionization chamber
(extended pencil, 30
cm lenght)
– But also OSL.
Thimble chamber
GafChromic
XR-QA
25.4 x 30.5 cm
300 mm
extended pencil
chamber
OSL
Characterisation of the x-ray beam
• Clinical implementation of dosimetry free in air.
Why not dose area product, like in radiography
and fluoroscopy?
Dosimetry in phantoms?
• CTDI is currently the most
widely used dosimetric
quantity in CT
• The concept of CTDI has
serious limitations (not
accurate, not additive, …)
• Larger phantoms (e.g. 35
cm long) are expensive and
HEAVY (body phantom
34.5 kg)
• The regular phantoms may
still be used, with some
correction factor
12
Monte Carlo dosimetry
• Monte Carlo dosimetry should be used for
development and assessment of new
dosimetric concepts in 256 channel CT
• Monte Carlo dosimetry is required for
calculation of patient dose: organ dose
and effective dose
13