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