CT 2

Axial CT

Gantry stops + rotates to get data from single slice, X-rays switched off, pt moves to next slice, Rotates to acquire data from next slice

Helical CT

AKA spiral/volume CT
Gantry rotating continuously releasing x-ray beams
table simultaneously moves
results in a continuous spiral scanning pattern

MDCT

Multi-Detector CT
2/more rows of parallel detector arrays
Allows acquisition of multiple slices in a single rotation

Advantages of MDCT

Faster scanning time due to wider total active detector width
Fewer motion artefacts

Reduced patient risk

Ideal for trauma imaging – cover entire pt in 1 scan
Fast scan times minimise time on table for critically ill patients
Paediatric scanning can be done with less sedation
Less contrast required reduces risk of adverse reaction

Thinner slices

Improved z-axis resolution
Isotropic imaging (equal voxel dimensions)
Improved multi-planar reformats (MPRs)
Improved 3D image rendering

MDCT detector

slice thickness is determined by collimation
Electronic detector selection (detector switching)

configuration

3 types:
Uniform/linear
Non-uniform/adaptive
Hybrid/mixed

Uniform

All rows have same size, width, thickness

Non-uniform

Not all equal
Smaller in middle, larger outside
Improves dose efficiency
Less division/ dead space
Expensive
Flexible

Hybrid

Set of narrow + set of non-uniform
Main type

Pitch

= ratio of distance moves per rotation to total w/ beam width

Higher pitch

less pt dose + quicker
Lower imaging quality because less images acquired

lower pitch

more pt dose + quicker better imaging quality

Beam pitch

able distance travelled in 1 360 by gantry rotation divided by total thickness of all simultaneous acquired slice.

Pitch determination

Pitch determined by how quickly table moves ---> in MDCT, factor in total thickness because there are more than 1 detector
e.g. in multi: 7.5/4 * 2.5= 0.75 while in single 7.5/5.0=1.5

cone beam acquisition

Having more detectors and more slices means a having wider beam width
A cone beam is required to cover the whole detector width

Cone Beam Reconstruction

With increased number of slices, the cone beam generates cone beam artefacts
As tube rotates, off-centre objects are visualised by different detector rows

Cone Beam Interpolation (2)

Tilted Reconstruction- produces non-axial images which are then filtered to produce standard axial images

‘Feldkamp Algorithm’ - a 3D back projection (standard FBP is planar + therefore 2D)

Tilted (Oblique) Reconstruction

Reconstruct using BP, at an angle to the axial plane
Overlap reconstructions and filter along the z-axis
Basis of GE and Siemens techniques

Feldkamp Algorithm

Measurements are being taken from different angles for each patient ‘voxel’.
The section of pt being imaged is divided into 3D voxels rather than a 2D matrix of pixels as happens in back projection

Image Quality in CT

image showing visibility of anatomical structures, various tissues, + signs of pathology
A measure of how suitable an image is for its intended diagnostic purpose
Suitability is determined if specific relevant criteria are met

Desired attributes

Good image Q but low dose
Low noise
Fast scanning
Free of artefact
High spatial resolution
Less blurring

Factors affecting CT

pt factor, reconstruction, scan parameters, viewing conditions, re solution

Noise

Variation in CT no. which isn't related to true attenuation co-efficient
Amount of ‘mottle’ in image

Noise occurs because...

Random variation in photons detection – stochastic noise
stat fluctuation in x-ray production /interaction/detection
Electronic noise=measuring system
Reconstruction noise

disadvantage of noise

Lower noise= better LCD (low contrast detection)
Smooth image does not vary from the value
Noise= can mask detail

Quantifying noise

Measure noise/deviation
Can be quantified: standard deviation in %

Factors affecting noise

Scanner specifications and design
Scanning acquisition parameters
Reconstruction parameters
Patient factors

Scanner specifications and design

Efficiency of detectors
X-ray beam filtration
Scanner geometry

Scanning acquisition parameters

Tube voltage
Tube current
Scan time
Slice thickness
Pitch

Reconstruction parameters

Back projection algorithms
Noise filters

Patient factors

pt size