CT

Question 1

What is CT dose index?

Answer 1

The CTDI is a measure of the absorbed dose from a single rotation of the CT scanner gantry, with no movement of the patient couch. It is usually expressed in milligray (mGy). CTDI is related to the absorbed dose.

Question 2

What is dose length product?

Answer 2

Dose-length product is simply the CTDIvol multiplied by the scanned length

DLP is proportional to the integrated absorbed dose and is useful for comparing doses between patients for the same examination.

measured in milligray centimetres (mGy cm).

Question 3

What is effective dose in CT?

Answer 3

Effective dose is used to describe doses to patients in terms of relative biological radiation risk for an examination and takes into account the radiation sensitivity of the different organs in the body. Effective dose can also be used to compare radiation risk with scans of different body regions or other types of radiological examination.

Recall that the effective dose (E) for a radiological examination is calculated from the sum of all the individual organ equivalent doses (HT), each weighted by the appropriate tissue weighting factor (wT).

The units of effective dose are millisieverts (mSv).

Effective doses in CT can be calculated using special computer programs. These are based on computational ‘Monte Carlo’ simulations of organ doses using a mathematical model of the body

Question 4

Other than kVp and mAs, what effects CTDI?

Answer 4

the peak dose and the shape of the dose profile depend on the CT scanner hardware configuration including Focal spot size
Collimator
Focus-isocentre distance (FID)

Question 5

Where is the dose highest in CT?

Answer 5

Because the x-ray beam is attenuated as it passes through the body, doses are highest at the periphery and lowest at the centre.

Question 6

What is CTDIw?

Answer 6

CTDIw is calculated as a weighted sum of the centre and average periphery measurements.

CTDIw is a good estimate of the average dose to the phantom at the central slice of a CT examination, as though it were scanned with contiguous slices over a distance of 100 mm in the z direction. It is expressed in mGy, the same units as CTDI.

Question 7

What is CTDIvol?

Answer 7

simply the CTDIw divided by the pitch, where pitch is the couch feed per rotation divided by the nominal collimation.

A larger pitch reduces dose while a smaller pitch increases dose, so CTDIvol is adjusted accordingly. CTDIvol is an approximation to the average absorbed dose within the volume of tissue that has been scanned. Again, it is expressed in mGy.

Question 8

UK-average effective doses for common CT examinations are:
Head (acute stroke) -
Chest (lung cancer) -
Abdomen (liver metastases) -

Answer 8

Head (acute stroke) 1.8 mSv
Chest (lung cancer) 14 mSv
Abdomen (liver metastases) 16 mSv

Question 9

How much greater is the effective dose of CT than planar radiography?

Answer 9

typically one to three orders of magnitude greater than planar radiography of the same body region

Question 10

How can sagittal or coronal slices be made?

Answer 10

interpolation from a stack of transverse slices.

Question 11

What physical property of the tissues do the pixel values in the x-ray CT image represent?

Answer 11

Linear attenuation coefficient

Question 12

What is the dominant physical process contributing to attenuation of x-rays?

Answer 12

Compton effect

The amount of Compton scatter depends on the corresponding component of the linear attenuation coefficient. This component is proportional to the electron density (the number of electrons per unit mass) of the material and, like the total coefficient, it is also proportional to the material’s physical density (mass per unit volume). Overall, therefore, it is proportional to the number of electrons per unit volume. For human tissues, the electron density is itself closely related to physical density.

Hints are provided below:

Pair production cannot occur for photon energies below 1.02 mega-electron volts.
The likelihood of photoelectric interactions decreases with increasing photon energy

Question 13

What is the CT number of air and water?

Answer 13

water is 0 and air is -1000

Question 14

How are CT numbers calculated?

Answer 14

Computed tomography numbers for other materials are calculated according to the difference in linear attenuation coefficient between that material and water

HU= 1000((LAC of tissue - LAC of water)/LAC of water)

Question 15

What is the difference in window and level?

Answer 15

Level is the middle number of the window. Window is how many levels are within above or below which it is displayed as white or black.

Question 16

How is data collected in a sequential CT scanner?

Answer 16

This type of scanner acquires only one slice at a time: after completion of data acquisition for one slice, the gantry is rotated back to its original position and the patient couch moved in preparation for the next slice.

To form a CT image, a series of projections through the patient are collected at discrete angles during rotation of the gantry around the patient.

Question 17

WHat is back-projection?

Answer 17

To reconstruct the attenuation values, a process called back-projection is used.

With no prior knowledge of the attenuation values, we assume that the row and column totals are made up of equal contributions from each of the boxes along a ray path. Therefore, to back-project the data, we evenly ‘smear out’ the attenuation totals back across an empty grid (or matrix) of boxes.

If we do this for both projections in our example and add the results together, we get a simple attenuation ‘image’. The attenuation values are then converted into CT numbers (pixel numbers) in the form of HUs.

Differences between the reconstructed and true values in this example are due to two factors:

Too few projections
Blurring caused by the back-projection technique

Question 18

how many projections are needed?

Answer 18

depend on many factors including scanner design, matrix size and reconstruction method. However, if too few projections are used to reconstruct a CT image then undesirable image artefacts (streaks) will result and image resolution may be degraded.

In clinical scanning, of the order of 1000 projections are typically acquired per rotation of the scanner gantry. The number of projections is not generally user-adjustable.

Question 19

What is filtered back projection?

Answer 19

To remove the blurring from the image, the projections are first combined with a special mathematical function called a filter (sometimes called a ‘kernel’). The combination process is called convolution.

There are two main effects on the projection:

Production of negative going edges at boundaries in the object. When back-projected, these tend to cancel the back-projection ‘streaks’ to give a sharper image on a uniform background
The amplification of noise in the projection
Back-projection of the filtered projections gives an image in which both the artefacts and the blurring are much reduced

Question 20

WHy is it important to choose the correct filter in FBP?

Answer 20

A range of filters are available on clinical CT scanners. In choosing the filter there is generally a trade-off between the sharpness of the image (good spatial resolution) and noise. It is therefore important to choose the correct filter for a specific diagnostic task.

Question 21

How are images required in helical CT?

Answer 21

Image reconstruction in helical CT involves a further complication because the patient couch is moved continuously during the scan and the gantry rotates continuously. Thus, the x-ray beam moves in a helical path along the body

This means that we must form our transverse slices from projections that are not only at different rotation angles, but also at different longitudinal positions along the scanner axis.

This is done by interpolating projections.

This differs from sequential CT, for which all the projections for a particular transverse slice are acquired within that slice.

Question 22

How are true axial slices reconstructed from single slice helical scans?

Answer 22

For reconstruction in single-slice helical scanning, the method used is 2 point interpolation. To generate a projection at a particular angle in the plane (slice) of interest, the two nearest projections at that angle on either side of the plane are combined in a weighted average.

Question 23

How are true axial slices reconstructed from multi slice helical scans?

Answer 23

Multi-slice CT uses an array of x-ray detectors arranged in multiple rows side-by-side. In multi-slice helical scanning, multiple projections are recorded at each angle of the spiral, one from each row of detectors.

filter interpolation is used. Typically, this is a fixed-width interpolation. All projections at a particular rotation angle, from any detector row and within a certain distance of the plane of interest, are combined in a weighted average.

The nature of the weighting function, or ‘filter’, may vary depending on application. For 16-slice scanners and above, more involved methods may be used, but the principle of filtering the image data over a fixed length remains the same.

Question 24

What is the purpose of filtering the beam in CT?

Answer 24

to remove low energy, or ‘soft’, x-rays, which would not penetrate through the patient, contributing only to radiation dose and not the image.

also narrows the energy spread of the x-ray beam making it more monochromatic. This is important in CT as the method of image reconstruction is based on the assumption of a single energy, monochromatic, beam.

Question 25

Why can a bow-tie filter be used?

Answer 25

Some scanners have a filter which has a shape that is wider at the edges than at the centre. When a view is acquired through the body, the x-rays traverse a greater thickness of tissue at the centre of the field of view than at the edges. Beam shaping filters attempt to compensate for this difference in attenuation across the field of view. This helps to reduce undesirable beam hardening artefacts in the image. Note that the bow-tie filter shapes the intensity profile of the transmitted x-ray beam, and not its physical dimensions.

Question 26

What do you think are useful functions of the collimator?

Answer 26

To restrict radiation dose to the patient - By restricting the x-ray beam to the active detector width, the collimator minimises unnecessary radiation reaching the patient that does not contribute to the image. The collimated width at the centre of the scanner (the isocentre) may vary between 1 and 40 mm To minimise scatter reaching the detector To define the imaged slice thickness

Question 27

What are the 2 different types of detector element?

Answer 27

Solid state Ionisation chamber

Question 28

What are important properties for a CT detector?

Answer 28

High detection efficiency for x-rays in CT energy range High dynamic range High geometric efficiency (narrow gaps between active elements) Fast response Linearity Stability Low cost Small physical size

Question 29

What makes up a solid state detector in CT?

Answer 29

cintillator material coupled to a high-purity, temperature-stabilised silicon photodiode to convert the scintillation light to an electrical signal. Scintillator materials are cadmium tungstate, bismuth germanate or doped rare-earth ceramics.

Question 30

What are the beneficial properties of SSDs?

Answer 30

High (~98%) detection efficiency ~80% geometrical efficiency Small physical size of detector elements A rapid light output with rapid decay time is desirable to allow rapid sampling of the signal as the gantry rotates around the patient.

Question 31

What makes up an ionisation chamber detector in CT?

Answer 31

Gas-filled ionisation chamber detectors are used in some single-slice scanners. These are filled with a high atomic number gas (e.g. krypton or xenon) at approximately 30 atmospheres pressure. The detector array is a single pressurised vessel, subdivided by tungsten septa, which also help to stop some scattered x-ray photons. Incident x-rays ionise the gas, producing a signal at the collection electrodes

Question 32

What enables continuous rotation of the CT scanner?

Answer 32

a 'slip ring' system where a series of 'brushes' on the rotating gantry maintain contact with stationary rings, allowing power to be supplied to the gantry and the detector signals to be passed out to the computer. Slip ring technology is essential for the technique of helical scanning

Question 33

What are the 2 ways in which CT scanners can acquire data?

Answer 33

In sequential scanning, also known as axial or 'step-and-shoot' scanning, the x-ray source is switched on and the gantry rotates to acquire data in a transverse (transaxial) slice while the patient couch is stationary In helical scanning, also known as spiral scanning, the couch movement and gantry rotation are continuous while the scan region is covered. Therefore, the x-ray beam describes a helical path relative to the patient

Question 34

What is pitch?

Answer 34

Pitch is defined as the couch travel per gantry rotation divided by the collimated beam width.

Question 35

What are the advantages of helical scanning?

Answer 35

Avoidance of respiratory misregistration as it can be performed in one breath hold More effective use of contrast agents - With faster scanning high vascular contrast can also be achieved Overlapping slices can be reconstructed from the helical data at any pitch without increase in radiation dose. Reduction in scan time and radiation dose

Question 36

Why can respiratory misregistration occur?

Answer 36

In sequential scanning requiring a breath hold, a small lesion can be missed if the depth of breath hold is different on successive slices.

Question 37

How are slices formed in multi slice scanning?

Answer 37

by a combination of beam collimation and electronic switching of the detector rows. For example, consider a detector array with 16 rows of 1.25 mm detector elements. By collimating the x-ray beam to cover half of each of the two innermost detector rows, a section width of 2 x 0.63 mm can be obtained.

Question 38

What are the different types of array in CT?

Answer 38

Linear array - A linear, or 'matrix', detector array has multiple rows of detectors, all of the same width. Adaptive array - In adaptive arrays, the detector row width is narrowest in the centre and increases outwards. Using fewer elements means fewer septa dividing the active area of the detector. This helps to improve geometric efficiency and therefore dose efficiency. Hybrid array - Hybrid arrays are similar to linear arrays, except that a central set of detector rows is narrower than the outer detector rows.

Question 39

What are the 2 different types of pitch in multi slice scanners?

Answer 39

Pitchx is defined as the couch travel per rotation divided by the total collimated beam width. Pitchd is defined as the couch travel per rotation divided by the detector subgroup width. For example, if a scanner is operating in 16 x 1 mm mode, the width of each detector row subgroup is 1 mm. With a couch movement of 24 mm per rotation, pitchx is 1.5 (24/16) while pitchd is 24 (24/1).

Question 40

What are the advantages of multislice scanning over single-slice scanning?

Answer 40

Faster scanning - total active detector width is generally greater than for single slice scanning Better dynamic imaging - As a consequence of faster scanning, it is possible to acquire scans during different phases of the passage of contrast media through the tissues of interest - for example, arterial, portal and interstitial phase liver imaging. Thinner slices - improving spatial resolution. Genuine 3-D imaging Simultaneous acquisition of multiple slices - This is associated with significant benefits - for example, the accurate determination of needle placement and angulation in biopsy procedures.

Question 41

What causes a ring artefact in CFT?

Answer 41

Detector sensitivity artefacts can occur if an element in the detector array is faulty or if the detector element gains are out of adjustment relative to each other. As the gantry rotates around the patient, the locus of the line of sight from the x-ray source to the faulty detector element describes a circle

Question 42

How do you avoid ring artefacts in CT?

Answer 42

To avoid ring artefacts, it is important to perform regular detector air calibrations to ensure individual detector element gains are correctly adjusted relative to each other. This is usually done each time the scanner is switched on, or on 'warm-up' after a period of inactivity.

Question 43

What is the trade off between thin partial voluming and noise?

Answer 43

To minimise partial volume effects, a slice thickness should be used that is less than or equal to the dimensions of the features of interest. However, thin slices are more noisy, so one must achieve a trade-off between the effects of excess noise and partial volume effect.

Question 44

What is partial intrusion?

Answer 44

Due to divergence of the x-ray beam between source and detector in the z-direction, an object may intrude into the slice at one projection angle, but not for the opposing projection. This creates inconsistencies between opposing projections, leading to streak-like artefacts in the final image

Question 45

What artefacts can arise from the cone shaped beam?

Answer 45

an off-isocentre small object may be detected by different detector rows in opposite projections leading to elongation and distortion

Question 46

What effect does beam hardening have on the image?

Answer 46

Linear attenuation coefficients decrease as x-ray energy increases. Therefore, attenuation coefficients (and hence CT numbers) will appear to be lower where beam hardening is greater, that is, through the centre of the patient. This leads to a 'cupping effect' in the CT numbers.

Question 47

How is the cupping effect corrected for in CT?

Answer 47

water calibration' can be performed on the scanner prior to scanning. This allows correction factors to be calculated and applied such that the image is uniform. The cupping artefact and its correction are clearer when the images are analysed to give CT number profiles across a horizontal diameter of the phantom Most scanners incorporate substantial pre-patient filtration to absorb soft x-rays and help minimise beam hardening effects. Some also employ 'bow-tie' shaped filters in an attempt to equalise attenuation across the patient profile

Question 48

Regarding x-ray beam hardening in CT: A. Low energy x-rays are attenuated less than high energy x-rays in the body B. After traversing the patient, the average x-ray beam energy is higher C. Beam hardening is the same throughout the FOV, for a given patient D. The 'cupping' effect, due to beam hardening can be corrected using air calibration data E. Appropriate beam filters, patient positioning and post-processing algorithms, can all help to minimise beam hardening artefacts

Answer 48

False True False False True

Question 49

When does photon starvation artefacts occur?

Answer 49

The photon starvation artefact can occur when some views (projections) are much more attenuating than others. For example, in the shoulders or pelvis, the x-ray path is much longer and penetrates through more bone in the lateral view than the anterior-posterior view. Therefore, a significantly lower photon flux reaches the detector in the lateral projections; the transmitted beam is literally starved of photons and so it has a higher noise content. This leads to a streak-like pattern of noise in the image, with the streaks oriented along the direction of greater attenuation.

Question 50

How can photon starvation artefacts be reduced?

Answer 50

mA modulation - the tube current (mA) can be varied as the gantry rotates around the patient. Higher mA is used for the more attenuating projections and lower mA for the less attenuating projections. The mA to be used is calculated either in advance, from analysis of the scout view or during the scan, via a feedback system from the detector. adaptive filtering where regions where the attenuation exceeds a specified level are identified and are subjected to additional smoothing before backprojecting

Question 51

Why does metal cause artefacts in CT?

Answer 51

Metal in the patient, due to its high absorption of x-rays, can cause localised beam hardening and photon starvation artefacts in CT images. These often result in streak artefacts.

Question 52

What artefacts can contrast media produce?

Answer 52

High concentrations of a contrast medium in vessels may, due to the effect of strong attenuation, produce streak artefacts in much the same way as for metal (although the artefacts are usually less severe).

Question 53

What measures can be taken to minimise motion artefacts in CT?

Answer 53

Helical scanning: helical scanning is less susceptible to errors caused by patient motion than sequential scanning Reducing scan time: the shorter the examination, the less likely the patient is to move. Faster rotation speeds and multislice scanning help to achieve this Breath holding: scanning during a breath hold may significantly improve many body scans. Good patient instruction and a manageable breath hold period are important factors Patient positioning: ensuring patient comfort and using appropriate immobilisation devices reduce the chance of motion artefacts Electrocardiogram (ECG) gating: for heart scans, using the ECG trace to trigger acquisition prospectively or reconstruct data retrospectively from specific phases in the cardiac cycle improve scan appearances

Question 54

What causes streak artefacts?

Answer 54

Inadequate FOV Photon starvation (inconsistencies are due to noise) Partial volume effect Motion Cone beam effects The presence of metal or contrast media

Question 55

What artefact is caused by a structure changing shape in helical scanning?

Answer 55

If the cross-sectional size or shape of an organ or body part changes in the longitudinal (z) direction, this can cause artefacts in helical CT. This leads to a form of partial volume averaging of the structure on either side of this plane. Due to the changing structure, this may lead to unsharpness and distortion.

Question 56

What artefact is caused by a structure changing position in helical scanning?

Answer 56

ifferent projections register the object as being in a different position. Geometric distortion of the object and inhomogeneities in the object and its surroundings may be produced.

Question 57

What type of artefact in sequential scanning is analogous to the artefact produced by a change in object position in helical scanning?

Answer 57

It is analagous to motion artefact. In both cases, the object is in a different position within the FOV for different projections used in the reconstruction.

Question 58

Certain aspects of CT scanner design place fundamental limits on transaxial resolution. Can you select three scanner factors affecting transaxial resolution from the list below? A. Helical versus sequential scanning B. Size of detector elements along a detector row C. Multi-slice versus single-slice scanner D. Focal spot size E. Focus-isocentre distance and focus-detector distance

Answer 58

A. Incorrect. B. Correct. C. Incorrect. D. Correct. E. Correct.

Question 59

why not just make the focal spot and the detectors extremely small to maximise resolution?

Answer 59

A greater number of smaller detector elements may reduce overall detection efficiency, due to the gaps between detectors. The smaller the x-ray tube focal spot, the more limited will be the x-ray tube power deliverable in a scan. Therefore, a careful balance in design is needed.

Question 60

What is quarter detector offset?

Answer 60

the centreline of the detector array is offset from the centre of rotation by one quarter width of a detector element. When the detector rotates by 180°, the opposing projection at the centreline is offset from that at 0° by half a detector element width, so that ray paths in opposite views are no longer coincident. This gives interleaved sampling of the patient by doubling the number of sampling rays per projection. Interleaved sampling gives a finer angular spacing of projections.

Question 61

What is the flying focal spot method?

Answer 61

the location of the focal spot on the anode of the x-ray tube is varied rapidly between two positions. This can be used to double the number of sampling ray paths per projection.

Question 62

The scan and reconstruction parameters can also significantly influence transaxial resolution. From the following list, can you select the three scan or reconstruction parameters which can influence transaxial resolution? A. Tube current milliamperage (mA) B. Number of projections C. Reconstruction filter D. Pixel size E. Tube kilovoltage (kV)

Answer 62

A. Incorrect. A change in the selected mA would not generally affect high-contrast resolution, except perhaps for very low mA where very high noise levels could affect the visibility of even high-contrast structures. B. Correct. A sufficient number of projections is essential to densely sample the object and produce good resolution. C. Correct. The reconstruction filter applied to the projection data may be varied to produce 'smooth' or 'sharp' images and has a significant influence on transaxial resolution. D. Correct. The pixel size sets a fundamental limit on resolution. E. Incorrect.

Question 63

How does the number of projections impact image resolution?

Answer 63

The angular sampling interval must be sufficiently small to allow detail in the object to be reproduced. improvement of spatial resolution as the number of projections increases

Question 64

How does reconstruction filter effect resolution?

Answer 64

Different reconstruction algorithms have different 'cut-offs' in terms of spatial frequency. The higher resolution, or 'sharp' kernels, for example those used to look at bony structures, have higher cut-offs while the lower resolution, or 'smooth', kernels have lower spatial frequency cut-offs. However, there is a noise penalty when sharp filters are used. Because noise is inherently of high spatial frequency, more noise is propagated through to the final image.

Question 65

How is pixel size determined in CT and how does this effect resolution?

Answer 65

Pixel size is determined by the choice of reconstructed field-of-view divided by reconstruction matrix. the highest spatial frequency that can be reliably reproduced is 1/(2xpixel) size

Question 66

What is targeted reconstruction?

Answer 66

One way to improve spatial resolution in a CT image is to re-reconstruct using a smaller field of view Unlike a simple zoom process in which the original image data are interpolated to fill in the extra pixels, reconstructing on a smaller field of view allows greater detail of information to be extracted from the original projections.

Question 67

Why is it important to match displayable spatial frequency and the reconstruction filter?

Answer 67

a mismatch between the filter selected and the displayable resolution. The high resolution of the filter is not exploited, yet its use will increase noise in the image. The mismatch could be overcome by using a smaller field-of-view or zoomed reconstruction.

Question 68

What is z-sensitivity?

Answer 68

Z-sensitivity refers to the effective imaged slice width. This is defined as:the full-width at half maximum (FWHM) of the slice sensitivity profile. Z-sensitivity represents the scan resolution in the z-direction, perpendicular to the transaxial plane, and determines the ability of a scan to reproduce fine detail in this direction. The ideal shape of the profile is rectangular, but it is usually rounded at the edges.

Question 69

Why are thinner slices more noisy?

Answer 69

because a narrower beam width means that fewer x-ray photons contribute to the projections. As a result, quantum noise in the data is increased and this propagates through the reconstruction process to the CT images.

Question 70

What primarily determines the z-sensitivity in single slice?

Answer 70

In sequential, single-slice scanning, the collimation primarily determines slice thickness, and hence z-sensitivity

Question 71

What primarily determines the z-sensitivity in multi slice?

Answer 71

the width of detector element grouping typically defines the minimum reconstructed slice thickness, and hence z-sensitivity

Question 72

What is the recommended overlap in helical scanning?

Answer 72

It is recommended that a 50% overlap is used for single-slice helical acquisitions.

Question 73

How does pitch effect z-sensitivity?

Answer 73

Use of a high helical pitch can degrade z-sensitivity by 'smearing out' the slice profile. This only affects z-sensitivity on single-slice scanners due to the limitations of the 2-point interpolation that is applied. he effect is not seen in multislice scanners because a more sophisticated interpolation method is employed.

Question 74

What is isotropic resolution?

Answer 74

Isotropic resolution refers to the resolution being the same in all directions. More specifically, resolution is the same perpendicular to the transaxial plane (z direction) as in the transaxial (x-y) plane.

Question 75

What is a voxel?

Answer 75

A voxel is each of an array of elements of volume that constitute a notional three-dimensional space, especially each of an array of discrete elements into which a representation of a three-dimensional object is divided.

Question 76

What are Advantages of isotropic resolution?

Answer 76

Reduced partial volume effect due to thinner slices Better multi-planar reformatting (MPR). Data can be viewed in any plane without significant loss of image quality. This can have advantages in terms of patient positioning, as scans can be acquired in a different plane to that used to review the data Improved volume rendering - isotropic resolution enables the generation of improved 3D representations of the data, such as surface shaded display, without the 'step' artefacts caused by thick slices

Question 77

What is the typical Z-sensitivity ?

Answer 77

0.3-10 mm

Question 78

What is contrast in CT?

Answer 78

Contrast refers to differences in brightness between regions of an image that correspond to anatomically or physiologically different parts of the body e.g. a lesion and surrounding soft tissue. In CT, the most appropriate definition of contrast is the difference in average CT number (Hounsfield units (HU)) between the lesion and the background

Question 79

What is noise in CT?

Answer 79

Noise refers to variations in image brightness that are unrelated to the structures being imaged. An important type of noise in CT and other x-ray imaging modalities is random noise, i.e. variations in brightness that result from natural randomness in processes such as x-ray production and radiation interactions. Random noise is also called stochastic noise or quantum mottle. It represents variation in CT number that is not directly related to true variation in linear attenuation coefficient and is most noticeable when the number of detected x-ray photons is low.

Question 80

What is contrast resolution in CT?

Answer 80

Contrast resolution is the minimum detectable contrast between a region of an image and its surrounding background. It depends on the size of the region and the noise. In general, CT has the ability to depict very small values of contrast, i.e. it has very good contrast resolution.

Question 81

How does noise effect CT?

Answer 81

degrades contrast resolution. In other words, noise impairs our ability to distinguish subtle differences in contrast between, or within, tissues which may exhibit changes of just a few HU.

Question 82

What is quantum noise?

Answer 82

Quantum noise arises due to the stochastic nature of the x-ray detection process. Due to the stochastic nature of the detection process, the number of photons detected will vary randomly from one detector to the next about the mean, or 'true', value These random variations follow Poisson statistics

Question 83

How can you reduce the effect of quantum noise?

Answer 83

the greater the number of detected photons, the smaller is the noise. More photons also generally implies an increase in radiation dose. doubling the mA or rotation time will double the photon flux to the detector for a given projection, and hence the number detected. Doubling the slice thickness doubles the width of detector used to form an image. Therefore, once again, twice the number of photons are detected. Photon flux, is also affected by tube kilovoltage (kV). The two are not directly proportional, however. In the CT kilovoltage range (about 120 kV) a +/-20 kV change will typically cause a +/-30-40% change in x-ray tube output and will therefore affect image noise accordingly.

Question 84

How can reconstruction kernels effect noise?

Answer 84

Different reconstruction kernels have different 'cut-offs' in terms of spatial frequency. The higher resolution, or 'sharp' kernels, such as those used to look at bony structures, have higher cut-offs while the lower resolution, or 'smooth', kernels have lower spatial frequency cut-offs. Because noise is inherently of high spatial frequency, more noise is propagated through to the final image when a sharp kernel is used.

Question 85

How does pitch effect noise in the image?

Answer 85

In single-slice scanning noise does not change significantly with changes in pitch. In multislice scanning 'fixed-width' interpolation of projections is used to form an image. Only projections within a specified width are used to form the image. As pitch is changed, the number of projections falling within the fixed width region changes. This means the amount of information used to form the image changes and therefore, so do noise levels.

Question 86

A number of factors can influence contrast between tissues in CT images. Can you select three of them from the following list? A. Tube current (mA) B. Inherent tissue properties C. Beam kilovoltage D. Reconstruction algorithm E. Use of contrast media

Answer 86

A. Incorrect. Tube current (mA) does not affect contrast, but it does affect noise and hence contrast resolution (the minimum detectable contrast). B. Correct. Inherent tissue properties (density, effective atomic number and electron density) can influence contrast between tissues in CT. Recall from the CT image formation session that pixel values in CT images are derived from the linear attenuation coefficients of tissue. For the x-ray energies used in CT, linear attenuation coefficient depends predominantly on tissue density. C. Correct. Beam kilovoltage (and hence beam energy) can influence contrast between tissues in CT. D. Incorrect. E. Correct. Use of contrast media can influence contrast between tissues in CT.

Question 87

What is the dominant source of noise in clinical CT images?

Answer 87

The dominant source of noise in clinical CT images is quantum noise.

Question 88

How are CTDI, DLP and effective dose related to mA?

Answer 88

Computed tomography dose index (CTDI), dose-length product (DLP) and effective dose are all proportional to the x-ray tube current (mA) i.e. for all other parameters constant, doubling the mA doubles the dose.

Question 89

How are CTDI, DLP and effective dose related to rotation time?

Answer 89

CTDI, DLP and effective dose are all proportional to the gantry rotation time. i.e. for all other parameters constant, doubling the rotation time doubles the dose.

Question 90

How are CTDI, DLP and effective dose related to pitch?

Answer 90

CTDI, DLP and effective dose are all inversely proportional to pitch. i.e. for all other parameters constant, doubling the pitch would half the dose.

Question 91

How does increasing kVp effect dose?

Answer 91

In CT, for all other parameters constant, an increase in kVp will result in an increase in patient dose. Increasing kVp results in a greater output from the x-ray tube. The efficiency of conversion of electron kinetic energy to x-rays at the anode target improves as kVp increases. There is an apparent contradiction here with planar radiography, where high kVp techniques are often thought of as low dose techniques. But remember, in planar radiography, the dose to the image receptor should remain approximately constant. Therefore, if higher kVp is used, the greater output and penetration mean a lower tube current should be used. However, in CT, the kVp can be increased without modifying the tube current. Image quality may improve due to less noise, but at the expense of increased dose.

Question 92

Suppose that the same region of a patient was imaged using a sequential (axial) scan protocol and a helical (spiral) scan protocol (Fig 1), all other parameters (kVp, mAs, imaged scan length, table feed per rotation etc.) being kept the same. Which protocol would give the higher dose?

Answer 92

For equivalent scan sections the dose for helical and sequential scanning would be about the same. Because helical reconstructions are formed by interpolating projections either side of a slice, a small amount of 'overscan' is required beyond either end of the imaged region in helical scanning. This causes a slight increase the helical dose, corresponding to 0.5-1 extra rotations at either end of the scan. However, each sequential rotation also includes some overscan, of 1-10%, in terms of rotation angle so that each rotation is in fact >360°. This helps, for example, with patient motion correction. Note, however, that helical overscan has a larger proportional effect for shorter scans and wider beam collimations. The dose increase due to overscan may be greater in multislice scanning. For short scan ranges, sequential scanning may therefore be more dose efficient than helical multislice scanning.

Question 93

How does dose change with collimated width in single slice scanning?

Answer 93

In general, for a single-slice scanner, dose to the patient does not change with slice thickness, i.e. collimated thickness. However, there is often an exception at the narrowest collimation. only the post-patient collimator is narrowed to the minimum width. The pre-patient collimator remains wider, ensuring alignment. However, this also means the patient receives some unused radiation. Relative doses can be up to 100% higher at 1 mm compared to other collimations.

Question 94

How does dose change with collimated width in multislice scanning?

Answer 94

For narrower collimations, the penumbra (overlap) constitutes a larger proportion of the x-ray beam. That is, a greater proportion of the beam is unused, meaning dose efficiency is reduced.

Question 95

Assuming that other factors are not changed, in a CT scan of the abdomen, effective dose will be reduced by: A. Increasing the gap between slices B. Increasing the mA C. Increasing the number of x-ray detectors D. Using helical scanning instead of sequential scanning E. Decreasing the field-of-view

Answer 95

A

Question 96

What is the HVL of tissue for CT roughly?

Answer 96

4 cm

Question 97

How can you adjust for patient size in CT?

Answer 97

Prior information - used by the operator to set the tube current (mA) for the examination according to standard tables. Weight or age-based protocols are particularly recommended in paediatric CT Scout view - information about cross-sectional attenuation can be calculated from the scout view, compared against reference values and used to set the tube current (mA).

Question 98

Why is mA tube modulation good?

Answer 98

mA can be changed for those views at which the diameter through the patient, hence attenuation, is less or more. Using rotational mA modulation, it is possible to reduce the overall mA, hence dose to the patient, without significant reduction of image quality.

Question 99

What type of scanning has the greatest dose efficiency generally?

Answer 99

A single-slice sequential scan has the greatest geometric dose efficiency because the total width of the divergent x-ray beam is always the same as or smaller than that of the row of detector elements. The total beam width includes the penumbra that is created by the pre-patient collimator due the finite size of the x-ray tube focal spot. This means that all the x-rays traversing the patient contribute to the images.