Imaging Flashcards
What is the aim of 4D adaptive radiotherapy
Continually adapt and reoptimise the plan (using IGRT information to establish tumour dose and OAR dose regularly)
What are the categories of IGRT
Simple (don’t require serial imaging, gross error check on fraction 1)
Complex (involve serial imaging for systematic error correction)
Adaptive (involving image acquisition so position of field can be changed at time of treatment)
What are the four types of dosimetric adaptation?
Scheduled ART
Reactive ART
Proactive ART
Real time ART
What is scheduled ART and an example
Replanning is scheduled in advance for predictable or likely changes
Pre-scheduled repeat scan is booked for H&N patients where weight loss is expected
What is reactive ART and an example
Acts on observed changes detected during imaging during treatment
Change in patient shape is observed and repeat planning CT is acquired
What is proactive ART and an example
Predicts changes likely to occur and prepares a choice of plans (library) to compensate
Bladder radiotherapy: create a number of plans with PTVs of different sizes for different bladder filling
What is real time ART and an example
Creates and delivers a new plan online while the patient is in the treatment position
VIEWRAY MRIdian
What do we need to consider for adaptive planning in general?
CTV includes sub clinical disease: how can we be sure we are still treating disease we cannot see?
What do we need to consider in reactive ART?
Is the dose to targets too high or low, is the dose to critical structures too high
Consider dosimetric, clinical, and patient specific factors
What are issues with plan of the day?
What about bladder filling during treatment?
Increased workload of 6 plans per patient - do you QA them all?
How many fractions do you allow of each plan?
How are you sure you are treating on intended plan?
What are the two main types of motion management?
Active and passive
Passive accepts that motion exists, attempts to quantify range of motion, creates treatment to incorporate movement prior to delivery
Active could reduce amount of motion, or use real time tracking, or adapt throughout treatment
What is an example of passive motion management?
Acquiring a 4D free breathing CT
Delineating ITV across 4DCT
Delivering treatment with free breathing
What are examples of active motion management?
Compression
Gated delivery or breath hold
Tracking
Considerations of compression
Treatment delivery is not interupted
Reproducibility may be a problem
Patient compliance may be an issue
Considerations of gated delivery or breath hold
Reduced duty cycle
Extended beam on time
Minimising breathing motion but increasing possibility of postural changes
Patient compliance
Considerations of using tracking
Beam on throughout
Relies on accurate prediction model
Relies on verification during delivery
Name some respiratory measuring systems
ANZAI
Pressure sensor in elastic belt around diaphragm
Varian RGSC/RPM
Small marker acts as surrogate of breathing motion using IR/visible light camera
C-RAD
Optical surface scanning with no fiducials or markers, can be used to assist patient posture correction
What is the most reproducible breath hold position?
Full exhale
How does the type of scan impact what is seen on CT?
CT will capture tumour at random point
PET will capture a time averaged tumour position
4DCT captures entire movement
Breath hold captures tumour at extreme position but is not going to have motion related artifacts and will have most true tumour shape and size
What is prospective gating?
CT acquisition is only gathered during a defined interval of breathing cycle (acquisition is gated)
What is retrospective gating and what are the main types?
Data is retrospectively binned into phases, overlapping spirals are acquired, slow CT scan
Bin size is independent of number of bins
Amplitude binning and phase binning
What are the drawbacks of retrospective gating?
Imaging dose is much higher than comparable 3DCT
Automatic dose control may not be available
Limitations on scan lengths
Imaging anatomy outside of CTs normal FoV may not be possible
Is amplitude or phase binning better?
Amplitude shows fewer artefacts and is more accurate
Phase is better for reconstruction of peaks
What are the challenges of using 4D images?
Single movie loop visualisation, is it representative of breathing cycle
Has the surrogate modified the breathing behaviour
What are some common 4DCT arefacts
Duplication of anatomy due to significant change in amplitude of breathing
Missing data - interpolated between points because no discernible points of data
‘Step’ artefact - system assumes linear breathing but patient is not breathing as expected
How does gated RT work?
Minimise motion by only treating in certain parts of breathing cycle
Acquire a 4DCT over the entire cycle, assess motion, select phases of 4DCT for planning to restrict motion
Reduces volume to be treated, but requires very regular breathing, audio coaching
What is the aim of abdominal compression
Restrict motion of the diaphragm and related respiratory motion
Note that frames alone are not sufficient for localisation: need IGRT
Why use DIBH
Reduces movement, creates more space between heart and breast
Patient holds breath for about 20 seconds, requires coaching, acquire 3DCT in BH
What imaging dataset do you want to match to a CBCT?
AvIP will have most similar appearance to CBCT, as CBCT averaged over many breathing cycles
Verify passive management with AvIP
How would you do real time verification and is it always appropriate?
Use 2D MV port during treatment
Not always appropriate: depends on location, tumour size, density which all impact contrast
How would you do verification when using compression devices?
Same as passive management because it is via physical restriction
Advantageous if TT are quick
How would you verify DIBH?
Compare reference breathing trace with measurement on day, should be consistent
Triggered 2D imaging in breath hold to verify and correct position
Patient visual feedback helps
How would you do verification for gating
Compare reference breathing trace with measurment on day
CBCT to set up anatomy and check motion
Trigered 2D imaging on entry and exit from gated window
Fluoro to assess tumour motion
What are some issues of treating bariatric patients?
Might be weight limits on machines, can they be treated/imaged?
Can patient be positioned so that there is no collision risk and so treatment is feasible? (not through arms eg)
Can the patient external be visualised?
Reduction in image quality - more attenuation and scatter in larger patients. Does protocol need to be changed for better image? Thicker slices? AEC running at capacity?
Do we need to assign densities where we can’t see patient properly?
How will verification be done?
How could we deal with not seeing patient external properly?
Ladder interlude - radiodense markers on patients surface make it very clear where external is
What could you consider verification wise for bariatric patients?
Increased frequency
Imaging in 3D if usual protocol is 2D
Increasing PTV margins
Change IGRT tolerances
What does imaging equipment need in order to be suitable for planning or verification?
Hard flat couch for reproducible position
Compatability with immobilisation equipment
Lasers for patient positioning
Geometric fidelity
(for planning need to be able to map to density)
How do we get dose from CT numbers?
CT number is converted to electron density or mass density via calibration
ED/MD values are used in dose inhomogeneity calculations
Why does HU vary with kV?
Linear attenuation is energy dependent, so lower energy will have lower relative difference in attenuation between materials, changing HU
Hu = 1000 x (mu - mu_water)/(mu_water - mu_air)
What are tolerances for CT QC on HU?
+/- 0.03 water, 0.05 lung, 0.08 bone
Calculate change in electron density which would produce 2% change in dose
What PET scans are often used and why?
FDG-PET
Tracer is glucose analogue, shows metabolic activity which is good for showing metabolically active tumours. Poorer spatial resolution
(PET-CT gives metabolic and anatomical information)
Why can PET be quantitative?
Because the attenuation along a (known) line of response is independent of the (unknown) position of the emission along that line
Why do we want to quantify PET images?
For lesion characterisation
Response assessment
Data reduction in trials, statistical analysis
Dose optimisation
Testing drug targetting
RT target identification
What is the SUV?
Standard uptake value
= activity concentration / (injected dose / body weight)
What is not standard about SUV?
Weight: could use body mass which is straightforward to assess but not best, lean body mass is more consistent but more complex to measure. Could use surface area, but difficult to measure
Definition of activity concentration, are you using SUV max (robust but in noisy images is skewed by noise and clinician may not find max), SUV mean (extremely dependent on contouring), SUV peak, proposed as compromise, calculates highest mean value of volume in ROI
What might you want to correct for in SUV?
Blood glucose level, correcting for this mans uptake values are more comparable. Error in blood glucose will propagate to SUV, so need proper lab testing
What is necessary to use absolute activity concentrations?
Scanner must be calibrated
Usually done via cross calibration to traceably calibrated dose calibrator to ensure consistency of activity measurements for SUV calc
Why might PET be useful in lung cancer?
Can discriminate tumour
Could also be consolidation (lung tissue that has filled with liquid instead of air) or atelectasis (collapse or closure of lung)
What is concomitant dose?
Dose within treatment course from sources other than treatment exposures, eg IGRT, CT, repeat CT
What would you consider when looking at DRLS?
Are protocols parameters difference (slice thickness etc)
Are the differences appropriate?
Do you need the improved image quality for advanced techniques?
Monitor per patient or retrospective audit?
Allows comparison but continually comparing and changing would result in continually worsening image quality
What should IGRT images aim to produce?
Doses as low as reasonably achievable
Adequate image quality for task
How might higher imaging dose protocol result in net dose saving?
Improved image quality could result in better target localisation or accuracy of treatment delivery
Eg can justify 4DCT as get better images of moving targets and can more accurately measure the motion
What can impact patient dose in CT/CBCT?
Irradiated volume - if much larger volume irradiated but without gaining more image, this is higher dose for same amount of image
How could CBCT dose be reduced?
Reduce scan length to reduce irradiated area - bigger impact if reducing exposure of more radiosensitive tissues - eg testes
Reduce mAs - scan phantom at different mAs and see when SD is acceptable at lowest mAs
What are appropriate units in SUV equation?
Activity concentration in Bq/ml
Weight in g
Dose in Bq
How is 3D IGRT usually carried out?
Have reference image and image taken on set
Compare images
Work out shift, apply shift
What methods of performing set up correction can be carried out?
On line, off line
Offline, take image but don’t do anything with it, make shifts afterwards
On line, take action based on image on the day
What is a gross error and what are some possible causes?
An unacceptably large set up error that could underdose the CTV or overdose an OAR
Not accounted for by CTV to PTV margin
Possible causes:
Incorrect patient, site, orientation
Incorrect field size, shape, orientation
Incorrect isocentre position of unacceptable magnitude
What is a systematic error, what types are there, and what are some possible causes?
A deviation that occurs in the same direction and is of similar magnitude for each fraction throughout treatment course
2 types: individual: mean error over treatment for one patient
Population: SD of distribution of mean errors for each individual patient
Causes:
Target delineation error
Target position and shape
Phantom transfer error
Patient set up error (tense at first plan and improves)
What is a random error, and what are some possible causes?
A deviation that can vary in direction and magnitude for each delivered treatment fraction
Individual (SD of errors over course of treatment)
Population (mean of individual random errors for each patient)
Possible causes: patient set up error
Target position and shape
Intrafraction errors
What is verification and what types are there?
Process by which geometric accuracy of RT is assessed
Pretreatment
Off line
On line
Real time
What are some margin calculation techniques?
CTV - PTV margin = a (sigma) + b (theta) + c
ICRU 62
Root((population systematic errr)^2) + (poulation random error)^2))
Stroem and Heijmen
2 x population systematic error + 0.7 x population random error
(ensures 99% of CTV receives more than or equal to 95% of prescribed dose)
Van Herk
2.5 x population systematic error + 0.7 x population random error
(ensures 90% patients in population receive minimum cumucative CTV dose of 95% prescribed dose)
