PHYSICS - CT Flashcards
Most modern CT scanners are what generation?
3rd generation (tube and detector spin in synchrony)
4th generation CT scanner
360 degree ring of DELs with rotating x-ray tube (no ring artifact)
Axial acquisiton
a.k.a. step-and-shoot; better spatial resolution in Z-direction (more projections per slice), no partial volume artifact in Z-axis
Helical acquisition
faster, flexible slice selection, reduced stair-step artifact, susceptible to partial volume artifact (due to interpolation), less discontinuity of moving anatomy between different slices (e.g. bowel)
Effective mAs
(mA * exposure time) / beam pitch, or mAs / beam pitch
Tube current modulation methods
based on topogram and/or on-the-fly
Anode-cathode axis relative to imaging plane
perpendicular (to decrease heel effect)
Slice width in SDCT
determined by beam collimation and equal to beam width
Slice width in MDCT
determined by DEL width and whether signal from adjacent rows are summed
Cone-shaped beam
3-dimensional, used in MDCT
Fan-shaped beam
2-dimensional, used in SDCT
Detector aperture size
a.k.a. DEL width (in Z-direction)
Determinants of in-plane spatial resolution
focal spot size, DEL size
Slice pitch
a.k.a. detector pitch; table movement per tube revolution / single DEL width
Collimator pitch
a.k.a. beam bitch; table movement per tube revolution / beam width
Relationship between pitch and spatial resolution
higher pitch => fewer projections => more interpolation required => lower Z-axis spatial resolution
Effect of a larger pitch
faster scan, lower dose, decreased spatial resolution, decreased SNR
Effect of a smaller pitch
longer scan, higher dose, increased spatial resolution, increased SNR
Pre-patient collimation
occurs in x/y and z directions; defines beam width and reduces overscan
Post-patient collimation
occurs only in z direction; scatter reduction
Geometric efficiency
% of beam that hits the detector; SDCT > MDCT with more rows > MDCT with fewer rows
Linear attenuation coefficient
describes attenuation per unit length of tissue; varies with kVp and tissue properties (Z and density); expressed in cm^-1; different for ice, water, and water vapor
Effect of increasing kVp on μ (LAC)
smaller μ (less beam attenuation per unit length of tissue)
Effect of increasing tissue Z on μ (LAC)
larger μ (more beam attenuation per unit length of tissue)
Effect of increasing tissue density on μ (LAC)
larger μ (more beam attenuation per unit length of tissue)
μ (LAC) for photon energies at the k-edge is increased or decreased?
increased μ (more beam attenuation per unit length of tissue at the k-edge)
Relationship between μ (LAC) and HVL
inversely related; materials with a higher μ have a smaller HVL
Mass attenuation coefficient
describes attenuation per unit mass of tissue; expressed in g^-1; same for ice, water, and water vapor
Water is always ___ HU
0 HU
Effect of reconstructing thicker slices from thinner slices
increased SNR, increased partial volume artifact, decreased spatial resolution in the Z-direction
Effect of a sharpening filter
increased spatial resolution, increased noise (decreased SNR)
Effect of a smoothing filter
decreased spatial resolution, decreased noise (increased SNR)
T/F - filters/kernals (in FBR) do not change voxel data
false; filters change voxel data (unlike window/level)
Advantage of iterative reconstruction
increased SNR at a given dose (or dose can be reduced while maintaining SNR), reduced streak artifact
180 degree axial interpolation (vs. 360 degree)
better spatial resolution, more accurate reconstructions, less partial volume artifact
Effect of decreasing pixel size on resolution and SNR
increased spatial resolution, decreased SNR (fewer photons per pixel)
Major factors affecting spatial resolution
pitch, slice thickness, DEL/pixel size, # of projections per slice, focal spot size, patient motion, reconstruction algorithm
Increased slice sensitivity profile (SSP) equates to…
decreased spatial resolution in the Z-direction; increased pitch => increased SSP => decreased resolution
SSP is increased by…
increasing pitch, using 360 interpolation (vs. 180), and detector binning => wider SSP => lower spatial resolution
How to: increase number of photons per voxel
increase slice thickness, increase DFOV, increase mAs, or decrease pitch; more photons per voxel => less noise
Effect of decreasing kVp
increased contrast, increased quantum mottle, decreased dose; may need to increase mA to maintain signal
How to: reduce dose
decrease kVp, decrease mAs, increase pitch, tube current modulation; also better reconstruction techniques
Cardiac imaging best performed during which part of the cardiac cycle
diastole
Prospective cardiac imaging
lower dose, always axial, no functional imaging, susceptible to motion (beta-blocker is essential)
Retrospective cardiac imaging
higher dose, helical (low pitch), functional imaging is possible, less susceptible to motion; use if CI to beta-blocker
Contraindication to beta-blocker - use prospective or retrospective gating?
retrospective
CT fluoro
lower mAs, same kVp; continuous and intermittent modes
Continuous mode CT fluoro
near real time; “first in, first out” image reconstruction
Advantages of dual source CT
faster acquistion, tissue characterization (with dual energy)
Contrast timing techniques
empirical, test bolus, bolus tracking
CT phantom sizes
16 cm for head, 32 cm for abdomen; contains central and peripheral pencil ionization chambers
Central and peripheral dose for head and abdomen
same for CT head; central dose is less than peripheral in CT abdomen
Dose (CTDIvol) is over or under-estimated for obese patients?
over-estimated
Dose (CTDIvol) is over or under-estimated for pediatric patients?
under-estimated
Adding doses together across multiple acquistions/phases
DLPs can be added together, while CTDIvol cannot
Z-axis variation
“tails” of radiation along the edge of the area being scanned; radiation profile is not limited to the primary area being scanned
Z-axis variation greater for SDCT or MDCT?
MDCT has greater Z-axis variation
How to: fix beam hardening artifact
tilt gantry or patient positioning, pre-harden beam (bowtie filter), calibration using a phantom, reconstruction algorithms; increasing kVp would also decrease beam hardening (per Ram)
How to: fix partial volume artifact
acquire thinner slices
How to: fix partial volume averaging
decreasing pitch, use thinner detector rows, use axial acquisition (instead of helical), reconstruct with thinner slices
How to: fix photon starvation
tube current modulation, adaptive filtration to correct attenuation profile
How to: fix undersampling
acquire as many projections as possible per rotation (for view aliasing) or utilize high resolution techniques (for ray aliasing)
How to: fix streak artifact from metal
thinner slices, higher kVp, remove metal (if possible), interpolation software
How to: fix patient motion
align scanner in the primary direction of motion, overscanning, gating, faster scanner, restrain patient
How to: fix incomplete projection artifact
position patient properly (e.g. move arms above head)
Ring artifact is specific to…
3rd generation CT scanners
How to: fix stair step artifact
thinner slices, reconstruct with overlapping slices
CTDIvol and DLP for typical head CT
60 mGy and 1000 mGy-cm
CTDIvol and DLP for typical abdomen/pelvis CT
15 mGy and 500 mGy-cm
Standard matrix size and bit depth
512 x 512, bit depth is 12
Overscanning
pitch <1
Effect of lower kVp on HU
lower kVp => higher HU, and vice-versa
Risk of fetal thyroid dysfunction with…
maternal IV contrast administration; considered a negligible risk
kVp for standard adult
120 kVp (range is 80-140); increased to 140 kVp for obese patients
Increasing number of photons 2x increases SNR by what factor?
1.4x
Changing kVp generally requires a change in…
mAs to maintain balance between dose and quantum mottle
mAs and kVp adjustments for pediatric imaging
decrease kVp (because thinner object) => need to increase mAs to avoid quantum mottle
Breast dose reduction strategies
reduce mA, tube current modulation, bismuth shield
Noise is proportional to…
1 / sqrt(number of photons per voxel); relates to mA, exposure time, and slice thickness
FBP vs. BP
FBP sharpens sinogram data prior to back projection; reduces streaking
Better SSP: 180 vs. 360 interpolation?
180 degree interpolation
Better SSP: axial vs. helical acquisition?
axial acquisition (no interpolation required)
Size specific dose index (SSDI)
SSDI = f * CTDIvol, where f >1 for obese and f <1 for peds
Sinogram axes
x-axis is DEL, y-axis is projection number