CTUG Nov 06
QA for MSCT - beyond IPEM 91
CTUG Nov 06
IPEM 91• IPEM Report 91 (2005)
– Recommended Standards for Routine Performance Testing of Diagnostic X-Ray Systems
• Chapter 12 CT– Image quality– CT number calibration– Radiation dose– Mechanical tests
– Axial, helical images– Inner and outer detector rows
CTUG Nov 06
Beyond IPEM 91
• All the slices?• MPRs• AEC• Cone beam artefacts• Other issues ?
CTUG Nov 06
• Four slice, outer slices noise ~ 5% high
All or some of the slices ?
0.000.050.100.150.200.250.300.350.40
1 2 3 4Detector Bank
Imag
e N
oise
(%)
CTUG Nov 06
• Noise 16 slice – not so predictable
All or some of the slices ?
0.80
1.20
1.60
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Slice #
noise
%
Std HeadStd Body
CTUG Nov 06
• Z-sensitivity (slice thickness) 16 slice
• QC is about change – does this graph matter ?• Test some or all ?
All or some of the slices ?
0.480.5
0.520.540.560.580.6
0.62
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Slice number
fwhm
(mm
)
CTUG Nov 06
Volume Imaging
• Volume imaging– MPRs, 3-D
• Should it be tested? – directly ?– indirectly?
CTUG Nov 06
MPR Direct Testing
• Noise
CTUG Nov 06
MPR Direct Testing
• Resolution– bead or wire orientated appropriately (PSF -> MTF)– Visual repeating pattern
0102030405060708090
100
0 2 4 6 8 10Frequency (lp/cm)
MTF
(%)
CTUG Nov 06
MPR Direct Testing
• Advantage– What you see is what you get– Takes into account any special reconstruction or
interpolation algorithms• Disadvantage
– MTF analysis, but fwhm PSF or visual can be fine
0102030405060708090
100
0 2 4 6 8 10Frequency (lp/cm)
MTF
(%)
CTUG Nov 06
MPR Indirect Testing
• Noise from helical slice
Recon position 1
CTUG Nov 06
MPR Indirect Testing
• 3-D resolution– z-axis (helical z-sensitivity)
0
50
100
150
200
-10 -8 -6 -4 -2 0 2 4 6 8 10
mm
CT
Num
bers
2.5mm5mm
FWTM
FWHM
perspexrod
Helical Z-Sensitivity tool
0.05 mm tungsten thin disk
CTUG Nov 06
MPR Indirect Testing
• 3-D resolution– x-y (scan plane)
0102030405060708090
100
0 2 4 6 8 10Frequency (lp/cm)
MTF
(%)
CTUG Nov 06
MPR Indirect Testing
• Advantage– Doing these tests anyway
• Disadvantage– Is it ok ? Recon is from raw data not from slices– But looking for change… – Doesn’t take into account special interpolation algorithms
in the 3-D
CTUG Nov 06
MPR Testing
• Should it be tested ?– Directly or indirectly ?
CTUG Nov 06
• Tube current modulation– Patient size, z-axis, rotational
– Axial and helical modes
Testing the AEC
From patient to patient Along patient length Around the patient
mA
angle-180 +180Low mA High mA
mA
CTUG Nov 06
• Test object to vary in z-axis and rotationally– eg Perspex phantom, conical with elliptical cross section
Testing the AEC
End view Side view
CT scanner couch
Catphan carrying case
• Based on ‘Apollo’ phantom developed by Muramatsu, National Cancer Centre, Tokyo
• Nick Keat – now at GSK
CTUG Nov 06
• Image along length of phantom – AEC off, on– Monitor image noise, mA, CTDIvol,
AEC off, Constant mA
Testing the AEC
CTUG Nov 06
• Circular, elliptical phantoms of various sizes– Scan short lengths over each section – Monitor image noise, mA, CTDIvol,
Testing the AEC
E. Castallano – RMH, London
CTUG Nov 06
Testing the AEC
GE LightSpeed16
0
4
8
12
16
20
24
28
-150 -100 -50 0 50 100 150Z-position (mm)
Noi
se (%
)
automA off
Noise Index 12
Increased mA
Decreased mA
www.impactscan.org/bluecover.htm
Increased mA
Decreased mA
CTUG Nov 06
Coronal view Sagittal view
z-axisAEC off
z-axis AEC on
Noise increases
Constant noise
Testing the AEC – Viewing with MPR
CTUG Nov 06
Testing the AEC
• Should this be a described test ?• In a specified phantom ?• How often?
CTUG Nov 06
Cone beam artefact
• Teflon (PTFE) rod in water, to simulate rib at an angle to scan plane
David Platten – now at Kings
CTUG Nov 06
Standard reconstruction AMPR
AMPR Algorithm
CTUG Nov 06
Windmill artefact in consecutive images
• Teflon (PTFE) rod in water, to simulate rib at an angle to scan plane (60°, Pitchx = 1.5, 16 x 1.5 mm acquisition, 5 mm image)
CTUG Nov 06
Testing the cone beam artefact
• Should this be a described test ?• In a specified phantom ?• How often?
CTUG Nov 06
Beyond IPEM 91
• All the slices ?• MPRs (+3-D), AEC, cone beam artefacts
– Should these be included in routine testing ?• Should there be anything else ?
0.480.5
0.520.540.560.58
0.60.62
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Slice number
fwhm
(mm
)
End view Side view
CT scanner couch
Catphan carrying case