CT principles Flashcards
CT scanners need:
- exact source-detector geometrical relationship
- stable and fast detecotrs
- high resolution
- mono-energetic beam
- tube perpendicular to fan beam
- computing power
- strong radiolucent bed
why does the tube need to be perpendicular to the fan bea,m
to avoid the anode heel effect
what is CT measured according to
- hounsfield scale
- according to the distribution of ‘u’
- u value being scaled to that of WATER
what is the CT number for water, air and bone
water = 0
air = -1000
bone = 1000
(everything else is somewhere in-between)
what is the number of different CT numbers that can be recognised by the scanner
2000
what is the image display format for most CT monitors
8 bit
256 different grey levels
how many grey levels can the eye differentiate compared to CT
Eye can resolve about 30 grey levels between black and white
CT scanner can differentiate 256 grey levels
how many CT numbers does soft tissue excluding fat cover
80
what is image reconstruction in CT
a mathematical process that generates tomographic images from X-ray projection data acquired at many different angles around the patient
what equation is used in CT image reconstruction
I = I0 e^-micro(x)
why are 2 views required in planar radiology
to give an idea of depth in an object
in CT what component provides depth of an object
grey scale
what is back projection and how is this done
- used to convert the data from the detector (projection space) to the image space.
- standard method of reconstructing CT slices by ‘smearing back’ the projection across the image at the angle it was acquired 9this reconstructs the image)
- done using tube input and pixel value
how can you make a simple back projection less streaky + why is it streaky in the first place
- the more projections added the better the representation, but this gives an increasingly streaky image
- hence filter (image processing) is applied
(filtered back projection)
why is back projection similar to the real picture but blurry
- because you ‘smeared’ the bright pixels across the entire image instead of putting them exactly where they are
how is filtered back projection done
- alter the projection data before back projection is done
- uses high pass filter or sharpening filter
what is the use of a high pass or sharpening filter in filtering back prjections
this type of filter picks up sharp edges within the projection
what is reconstruction kernel
how ‘soft’ the filter used in filtering back projection is
why is reconstruction kernel important
- high pass/sharpening filter can reduce blur BUT this accentuates noise as noise looks like jagged edges
- so using high pass filters make images look too grainy and softer filters are required (reconstruction kernel)
how would you adjust between kernels for scans
different kernels chosen for different images needed
e.g if u need to spot fractures, small discrete features, use hard kernel to accentuate these features
or
for soft tissue e.g brainy want to look for larger features with mild differences so u use softer kernel to decrease noise
what is iterative image recontruction
- takes an initial guess and refines over several iterations
- starts with filtered back projection then stimulates making CT reconstruction of initial guess
- by looking at difference in reconstruction, you can make better guess
- process is repeated for defined number of cycles
why does iterative image reconstruction perform better in high-noise situations e.g low dose scans
simulation used by iterative reconstruction technique attempt to model real-world noise and attenuation process
what is the main negative of using iterative image reconstruction
much longer reconstruction time and image have different subjective quality than filtered back projection
what does CTDI stand for/what is it
computed tomography dose index
- distribution of radiation dose in a single CT scan
ideally, radiation dose would drop sharply at edges of scan but this does not happen, what does? what does this cause and why does it happen
how can the radiation dose of slices of tissue be affect because the radiation dose does not drop at edge of scans
- radiation dose profile process gradually at edges of scan with penumbra on either side
- these tails/penubra is due to both xray beam divergence and internal radiation scatter by body tissue
- so overall, each slice of tissue in CT not only receives radiation as it gets scanned but also gets some from when adjacent slices are scanned
what is the integration length of CT limited to
100mm
what is CTDI 100
A measure of the dose over a pencil chamber 100mm in length
what does NT define
number of detector rows x width of detector row (irradiated slice width)