Fluoroscopy & DSA

Question 1

What is the range of frame rates in fluoroscopy and what limits this?

Answer 1

3-30 fps. This is limited by the how many fps the viewing monitor can display - acquiring at a frame rate higher than this would contribute to needless patient dose.

Question 2

What are the distinguishing features of a cardio lab fluoroscopy unit?

Answer 2

• Reasonably powerful tube with good cooling - some cardio procedures can last a long time.
• Small flat panel detector area - only small FOVs are required in cardio procedures.
• Detector optimised for speed (less image lag) and resolution (no pixel binning) - procedures require precise positioning of small catheters etc.

Question 3

What are the distinguishing features of an angiography fluoroscopy unit?

Answer 3

• Powerful tube with good cooling - some angio procedures can last a long time.
• Large flat panel detector area for a range of FOVs.
• Usually capable of CBCT (i.e. typically acquiring images during a ~ 1 s half or full rotation around patient).

Question 4

What are the distinguishing features of a general fluoroscopy room?

Answer 4

• Mid-range tube power covering a range of potential procedures (but not much longer/more complex procedures).
• Large flat panel detector area for a range of FOVs covering multi-purpose use.
• Cheaper than systems dedicated to more complex procedures.

Question 5

What are the distinguishing features of mobile fluoroscopy units? How are mini C-arms different?

Answer 5

• Flat panel or II.
• Portable.
• Mini C-arms have a lower power x-ray tube and a smaller detector area (suitable for extremity investigations).

Question 6

List some advantages of flat panel detectors over IIs.

Answer 6

• Improved dynamic range.
• Less image lag.
• No geometric distortion.
• Not affected by magnetic fields.

Question 7

What is the effect of reducing field size in an II?

Answer 7

• Electrons optics altered to map a smaller area of the input phosphor to the same area output phosphor.
• Magnifies image and improves spatial resolution.
• Reduces geometric distortion.
• Reduces minification gain and, therefore, brightness gain. This means an increased dose is required to maintain SNR.

Question 8

Explain the process of magnification for small flat panel detectors.

Answer 8

• For smaller flat panels, digital magnification is typically used. This involves simply increasing the pixel size.
• This corresponds to no improvement in resolution.
• Increased magnification will mean an increased dose is required to maintain SNR. This is due to the fact that the number of photons per unit area on the screen will be less due to the increased in pixel size.

Question 9

Explain the process of pixel binning.

Answer 9

• Pixel binning involves averaging multiple input pixel values into bins of larger output pixels.
• Increased pixel binning corresponds to a degradation in resolution.
• It also improves SNR as there will be more photons per pixel.

Question 10

Explain the process of magnification for large flat panel detectors.

Answer 10

• For larger flat panel detectors, magnification typically involves variations in pixel binning.
• For larger field sizes, automatic collimation is set at a maximum (i.e. at the edges of the detector to ensure the x-ray field does not exceed the detector area).
• Increased pixel binning will be applied, thus improving SNR but degrading resolution.
• For smaller field sizes, automatic collimation is set to match the field size to ensure no unnecessary patient dose.
• The level of pixel binning will be reduced, thus improving resolution. However, increased dose will be required to maintain SNR.

Question 11

Explain how fluoroscopy ABC/AECs work. How is this different for older IIs and more modern IIs/flat panels?

Answer 11

• The ABC/AECs monitor image properties and automatically adjust exposure parameters to maintain a constant image quality.
• A specific pixel area can be monitored in digital systems to focus on a region of interest.
• Older II systems monitor the brightness at the CsI input phosphor or at the output phosphor and alter exposure parameters to keep this constant.
• Newer IIs/flat panels monitor pixel values in the raw image and alter exposure parameters to keep this or SNR constant.

Question 12

List some factors that can affect SNR/brightness in fluoroscopy.

Answer 12

• Exposure parameters.
• Filtration.
• Collimation.
• Setup geometry (i.e. focus-to-detector distance, patient positioning and C-arm orientation).
• Patient size.

Question 13

How is input dose rate altered by the ABC/AECs in fluoroscopy to maintain constant ouput brightness/SNR?

Answer 13

One or a combination of the following parameters can be altered (most systems use a combination of kV, mA and ms):
- kV - This will affect beam penetration and, therefore, the number of photons reaching the detector.
- mA (tube current) - This will affect the number of photons produced per pulse and, therefore, the amount reaching the detector.
- ms (pulse width) - This will affect the length of a pulse and, therefore, the number of photons per pulse and the amount reaching the detector.
- pps (pulses per second) - Can increase the dose rate at the detector. However, not typically altered by ABC/AECs (usually user-selected).
- Filtration - Automatic filtration changes will affect beam penetration and, therefore, the number of photons reaching the detector.

Question 14

What is the purpose of fluoroscopy power curves?

Answer 14

Different fluoroscopy power curves alter the way the ABC/AECs vary kV, mA and ms and to what extent. Different power curves will be optimised for different purposes (e.g. kV variations may be limited in applications in which the kV needs to be matched to there K-edge of a contrast medium).

Question 15

What type of noise should a detector be limited by?

Answer 15

Detectors should be quantum limited (i.e. limited by the Poisson noise at the input). If not, optimisation techniques based on variations in dose may not work.

Question 16

How does an anti-scatter grid work? What effect does the use of an anti-scatter grid have on image quality and patient dose? When may an anti-scatter grid not be used in fluoroscopy? How else can scatter be reduced in fluoroscopy?

Answer 16

• Anti-scatter grids preferentially remove scattered photons by absorbing those incident at increased angles.
• This improves image quality as scattered photons contain no spatial information and, therefore, contribute to an increase in random background, thus reducing SNR.
• As an anti-scatter grid also attenuates primary photons, an increase in dose (~x2) is required to maintain SNR.
• => May be removed situations where reduced dose is required and the loss of contrast can be tolerated (e.g. in cardiology where procedures are long but high contrast objects such as wires may be used or in paediatrics).
• Scatter can also be reduced by collimating down to the area of interest as much as possible.

Question 17

What is the difference between fluoroscopy and fluorography?

Answer 17

Fluoroscopy:
- Typically involves lower dose rates.
- Lower image quality due to the lower dose rate.
- ‘Live’ imaging which is not usually automatically saved.
- Typically fixed ABC/AEC area.
- Usually used for positioning in interventional procedures or prior to diagnostic image acquisition.
Fluorography:
- Typically involves higher dose rates.
- Better image quality due to the increased dose rate.
- Images are usually automatically saved.
- ABC/AEC area can usually be altered.
- Typically used for recording high quality images.

Question 18

What is frame averaging and what are its effects? What is dynamic noise reduction/motion detection?

Answer 18

• Frame averaging involves averaging together fluoroscopy or fluorography frames (number of frames per unit time that are averaged can be varied).
• This reduces noise (meaning dose can be reduced) but also increases image lag (i.e. blurring off fast moving parts).
• Dynamic noise reduction/motion detection: Frame averaging of only non-moving areas of image to avoid image lag issues discussed above.

Question 19

What is one of the main dose concerns with regards to fluoroscopy procedures?

Answer 19

• Skin dose. Backscatter from tissue means the skin dose is ~ 30% higher than the incident air kerma. This can cause deterministic tissue reactions (ranging from erythema to necrosis) when the skin dose reaches ~ 2 Gy.
• Stochastic effects (cancer induction) are unlikely.

Question 20

List some factors that could affect patient dose in fluoroscopy.

Answer 20

• Screening time.
• Patient size.
• Collimation.
• Fluoroscopy power curve selection.
• pps/fps.
• Setup geometry (e.g. C-arm orientation and patient positioning).
• Field size.
• Anti-scatter grid.
• Contrast agents.
• kV/filtration.
• QA/optimisation.

Question 21

Increased screening time increases patient dose. How does screening time affect patient dose in fluoroscopy?

Answer 21

Fluoroscopy procedure times can be variable depending on the application (e.g. some interventional cardiac procedures such can be long). They can also vary from patient to patient depending on the complexity of the case.

Question 22

How does patient size affect patient dose in fluoroscopy?

Answer 22

Larger patients will receive an increased dose due to the increased exposure factors required by the ABC/AECs to penetrate the patient and achieve the required detector dose. An increased dose will also be apparent due to the increased scatter in larger patients.

Question 23

How does collimation affect patient dose in fluoroscopy?

Answer 23

Collimating down to the area of interest will reduce patient dose arising from direct and scattered radiation.

Question 24

How does fluoroscopy power curve selection affect patient dose?

Answer 24

Fluoroscopy power curves determine how the ABC/AECs vary exposure factors and, therefore, will directly affect patient dose. The selected curve should be optimised to the procedure.

Question 25

How does the pulses per second (pps)/frames per second (fps) affect patient dose in fluoroscopy?

Answer 25

Reducing pps/fps will reduce dose. However, pps/fps should be high enough for procedure requirements. It should be noted that reduced pps/fps does not always correspond to reduced dose on all systems; ABC/AECs may compensate for this reduction in dose; Some systems may change fps but not pps; Some systems may change the duty cycle (i.e. beam on/beam off time).

Question 26

How does the selected field size affect patient dose in fluoroscopy?

Answer 26

Using larger field sizes reduces skin dose due to the larger SNR apparent. This is caused by the increased minification gain/brightness gain in II systems, the photons per unit area (as seen on the screen) is lower in small flat panel systems due to digital magnification and the increased pixel binning meaning more photons per pixel in larger flat panel systems.

Question 27

How does setup geometry affect patient dose in fluoroscopy?

Answer 27

• C-arm angle can be varied to irradiate different areas of skin in some longer procedures. This will spread the skin dose, reducing the potential for deterministic tissue reactions.
• Positioning the patient closer to the tube increases patient dose due to inverse-square law effects. Therefore, patients should be positioned as close to the detector as is practicable.

Question 28

How does the anti-scatter grid affect patient dose in fluoroscopy?

Answer 28

Using an anti-scatter grid will increase patient dose due to the fact it attenuates primary photons as well as scattered photons. Therefore, removing the anti-scatter grid will reduce patient dose. However, this must be in procedures where the corresponding reduced contrast and increased noise can be withstood (e.g. paediatric work and some cardiac work).

Question 29

How do contrast agents affect patient dose in fluoroscopy?

Answer 29

Contrast agents attenuate more than surrounding tissues due to K-edge absorption, showing up better in the resultant image. They can be used to enhance contrast in areas of interest and, therefore reduce the required dose.

Question 30

How do kV/filtration variations affect patient dose in fluoroscopy?

Answer 30

kV:
- In contrast studies, the fluoroscopy power curve used will maintain kV at/just above the K-edge to maximise absorption and contrast.
- Greater kV can be used in low dose studies (e.g. in paediatrics). Penetration to the AEC will increase and, therefore, skin dose will reduce. However, there is a trade-off, where increased kV will reduce contrast.
Filtration:
- Can remove lower energy photons, reducing skin dose.
- For contrast studies, the above phenomenon can be used to preferentially remove photons below the contrast medium K-edge.

Question 31

How does equipment QA affect patient dose in fluoroscopy?

Answer 31

Routine QA monitors the dose and image quality associated with the system. Results can be used to guide optimisation of the equipment to minimise patient dose and give adequate image quality.

Question 32

How is patient dose monitored during and after a fluoroscopy examination? How does this contribute to reduced patient dose?

Answer 32

• DAP meters fitted to the tube can be used to monitor DAP and compare to expected levels for each exam.
• Newer systems estimate skin dose rate by correcting the measured DAP to some specified reference point. It should be noted that this often does not account for backscatter.
• Fluoroscopy screening times were used to monitor dose in the past. However, this does not truly reflect dose.
• These monitoring parameters can be used by clinicians in real-time to ensure patient dose is kept ALARP.
• Records of these monitoring parameters can also be audited subsequently to compare against local and national DRLs.

Question 33

List the steps of digital subtraction angiography (DSA) procedure.

Answer 33

• A non-contrast, ‘mask’ image is initially taken.
• A series of ‘live’ contrast images are then taken after contrast injection.
• The mask is subtracted from the ‘live’ images pixel by pixel, thus enhancing the contrast-filled regions

Question 34

What are some issues associated with digital subtraction angiography (DSA)?

Answer 34

• Increased dose compared to fluorographic acquisitions.
• Increased potential for motion artefacts.

Question 35

Assume a region of length T and linear attenuation coefficient mu_T for a digital subtraction angiography (DSA) mask image. Assume a smaller contrast region of length c and linear attenuation coefficient mu_c within the length T once contrast has been injected for a live image. Incident x-ray intensity is the same (I_0) for both the mask and live images. Exit intensity is I_1 for the mask image and I_2 for the live image. How are the two images subtracted mathematically? What does this subtraction assume?

Answer 35

• Mask image: I_1 = I_0.e^(-mu_T.T).
• Live image: I_2 = I_0.e^(-mu_T.(T-c)).e^(-mu_c.c)).
• Standard subtraction leaves a dependence on T and, therefore, background anatomy so is not sufficient. Logarithmic subtraction is required.
• Mask image: ln(I_1) = ln(I_0) - mu_T.T.
• Live image: ln(I_2) = ln(I_0) - mu_T.(T-c) - mu_c.c.
• Subtraction: ln(I_1) - ln(I_2) = mu_c.c - mu_T.c = c(mu_c - mu_T).
• This subtraction assumes a monoenergetic beam which is not the case.

Question 36

What happens to noise in a digital subtraction angiography (DSA) image when considering the mask and live images? What are the implications of this?

Answer 36

• As there is random noise that is different in each image, the subtraction process will increase the level of noise in the resultant image (noises will be added in quadrature).
• This means dose rate has to be increased to maintain SNR (although not dramatically as the image is contrast enhanced). Angio rooms tend to have a more powerful x-ray tube.