ELogs/JamesGolbourn

Electronic Log for James Golbourn

Project Direction

The project will be split into two sections, the first will be similar to the research essay and will be a survey on current quality assurance procedures clinically used. The second will be on focused on the features of an ideal integrated QA device, including what beam characteristics it will measure, how long it will take and the precision of the measurements. Potential basic CAD model of an ideal detector device e.g one that could be mounted onto the nozzle. When talking about ideal should be specific regarding which measurements, how quickly

Scintillator Columns

Raffy and I briefly discussed a method for QA measurements which involves stacking scintillator screens and then splitting the screens into columns such that depth and transverse profile measurements can be taken. The beam would be centred on the column and the spot position will be a measure of the deviation of the beam away from the predicted position.

File:Scintillator stack.png.

Potential QA with Sun QA 3

Would be good to make use of both the diode sets and the ionization chambers. To get range measurements at different depths could put a plastic build up in front of the electron ionization chambers corresponding to different depths. Spot position can be inferred from the diode triplet set. The shape and size of the spot can be reconstructed from the diode information e.g reconstructing Gaussian shape. Different spot patterns can be

Other Factors to Consider When Designing

A daily QA device will need to be robust enough to withstand knocks without altering the calibration of the device. There are also interesting applications for how the data collected by the QA device can be communicated with the user. For example, Simon and I discussed the possibility of the device being WiFi enabled and connected to a tablet device where the results from the QA procedure can be visualised on the tablet in real time. An alternative method would be for there to be a USB port on the device such that the data stored can be uploaded onto a computer screen and the QA results displayed on there. The storing of the data collected from daily QA has advantages. One advantage would be that direct comparisons could be made between that day's measurement and the immediately prior days, and if the recorded value lies within an error range of previous measurements this could provide an opportunity to reduce the time taken for QA procedures by comparing relative measurements. Furthermore, the storing of daily QA data would allow for a continuous tracking in the performance of the beam delivery system at regular intervals, which would better allow engineers to predict when maintenance work would be required on the delivery system. For visual measurements, for example those carried out with a scintillator screen to check for spot position, direct comparisons with previous measurements would be more difficult. Initially when a centre begins treatment after commissioning, storing the data collected is especially important to increase confidence in the delivery system and in QA measurements. The data can be used to inform which devices are required for different QA checks and to the precision required as the confidence in the stability of the system increases.

Alison and I also discussed further factors to consider when designing. Firstly, the weight of the detector, ideally want it to be light enough for one person to carry it. Scintillator screen currently used at UCLH is heavy and requires 2 people. The size of the detector is also important, smaller detectors are more practical but are often more expensive due to smaller electrical components. Alison thought that mounting the detector to the nozzle for measurements at multiple gantry angles would be ideal. Also interesting consideration more with commissioning, need to account for Earth's magnetic field with gantry angle due to scanning magnets. UCLH will initially want all data recorded from daily QA to keep a log of it.

To Do

Short Term

• Start to think about project direction.

Long Term

• Get in contact with Laurent contact at Med Austrom about QA procedures.
• Prepare practice presentation for project (8/2/19)
• Write up report on meeting with UCLH.
• Look at what functioning centres are using as QA.
• Look at advantages and disadvantages of QA systems.

Research Essay Device Summary

Sun Nuclear QA 3

Property measured	Component	Description	Precision
Output	CAX	The CAX is located at the centre of the SOBP and output is determined from charge collected in CAX. This is compared to a reference value.	2-3  %
Range	Electron Ionization Chambers	Top left and bottom right chambers are positioned in distal fall off region, top right and bottom left are positioned in SOBP centre using range compensators of different thickness to modify beam range. Ratio of charge measured in chambers is used to infer beam range fluctuations.	1 mm
Symmetry	Electron Ionization Chambers	Charge ratio between top right and bottom left chambers.
Spot position	4 diode triplet sets	Central diode of each diode was exposed to a dose at 50 % lateral fall off and the position determined by signal measured in neighbouring two diodes.	1.5 mm

Links to papers

1. Xiaoning Ding et al. “A novel daily QA system for proton therapy”. In:Journal of applied clinical medical physics 14.2 (2013), pp. 115–126. https://aapm.onlinelibrary.wiley.com/doi/full/10.1120/jacmp.v14i2.4058
2. James E. Younkin et al. “Technical Note: An efficient daily QA procedure forproton pencil beam scanning”. In: Medical Physics 45.3 (2018), pp. 1040–1049. doi: 10.1002/mp.12787. https://www.ncbi.nlm.nih.gov/pubmed/29394447
3. Lambert, Jamil, C. Bäumer, Benjamin Koska and Xiaoning Ding. “Daily QA in proton therapy using a single commercially available detector.” Journal of applied clinical medical physics (2014). https://www.ncbi.nlm.nih.gov/pubmed/25493526
4. Sun Nuclear QA 3 fact sheet: https://www.sunnuclear.com/documents/datasheets/dailyqa3-rfdaily.pdf

IBA Matrixx

IBA Matrixx formed of 1020 parallel plate ionization chambers arranged in a 2D array. There are 2 versions used in PBT, the Matrixx PT and Matrixx Evolution. The PT is more precise than the Evolution with the chamber spacing improving from 5 mm in the Evolution to 2 mm in the PT this improves the resolution. PT is more sensitive (1.3 nC/ Gy) than the Evolution (2.4 nC/Gy).

There are large dose uncertainties in the Evolution, particularly in the penumbra regions of upto 4 %, this is improved with the PT device to 1 % primarily due to the improved resolution. The Matrixx is a useful device as it provides results in real time and has been demonstrated to have a similar precision to film dosimetry techniques.

Evolution

Property measured	Component	Description	Tolerance
Output	3 x 3 array of chambers	Area exposed by beam and dose measured from charge in ionization chambers. Uncertainty is reduced by averaging over several chambers. Plastic water was placed in front of chambers to make sure chambers were at centre of modulations. Value compared to farmer ionization chamber reference.	1 %
Flatness and Symmetry	Ionization Chambers	Flatness determined by how the mean dose varied across the central 80% of the full width half maximum of the beam whilst the symmetry was determined from the difference in the largest points.	1 %

Links to Papers

1. Bijan Arjomandy et al. “Use of a two-dimensional ionization chamber array for proton therapy beam quality assurance”. In: Medical Physics 35.9 (2008), pp. 3889–3894. https://www.ncbi.nlm.nih.gov/pubmed/18841839
2. Liyong Lin et al. “Use of a novel two-dimensional ionization chamber array for pencil beam scanning proton therapy beam quality assurance”. In: Journal of Applied Clinical Medical Physics 16.3 (2015), pp. 270–276. doi:10.1120/jacmp.v16i3.5323. https://europepmc.org/articles/pmc5690130
3. M. Varasteh Anvar et al. “Quality assurance of carbon ion and proton beams: A feasibility study for using the 2D MatriXX detector”. In: Physica Medica 32.6 (2016), pp. 831–837. issn: 1120-1797. doi: https://doi.org/10.1016/j.ejmp.2016.05.058. https://www.ncbi.nlm.nih.gov/pubmed/27246359
4. Matrixx Evolution and FFF factsheet: https://www.iba-dosimetry.com/product/matrixx-universal-detector-array/

IBA Lynx and Sphinx

Scintillator based device. High spatial resolution of 0.5 mm and a large active surface area of 300 mm x 300 mm. The CCD pixels have a physical size of 4.65 µm x 4.65 µm. The Sphinx is designed to compliment the Lynx and is made up of blocks of RW3 of varying thickness and shape (wedged and non wedged). By scanning a beam across the block range measurements can be performed. The Sphinx is mechanically bolted to the Lynx and a parallel plate ionization chamber is contained in one of the RW3 blocks for output measurements.

The Lynx was irradiated with a uniform spot pattern consisting of 9 identical spots in a 3 x 3 array with each spot separated by 100 mm. The centre of the spot was calculated as the coordinate of the midpoint between the 50% levels on the beam profiles. The detector was irradiated for an exposure time of 300 s.

Property measured	Component	Description	Precision
Spot Size	Diode	The spot size was determined from the full width half maximum of the spot	2 %
Spot Position	Diode	spot position was measured as the deviation between the measured and expected position of the spot.	0.5 mm
Range	Sphinx blocks and diodes.	The beam is directed along the wedge to obtain the depth-dose curve and from this, range characteristics such as the proximal and distal range were obtained.	2 mm
Output	Ionization chamber.	No method described but could focus beam on block containing chamber and compare to reference chamber.

Links to Papers

1. S. Russo et al. “Characterization of a commercial scintillation detector for 2-D dosimetry in scanned proton and carbon ion beams”. In: Physica Medica 34 (2017), pp. 48–54. issn: 1120-1797. https://www.ncbi.nlm.nih.gov/pubmed/28118950
2. Lorenzo Placidi et al. “Range resolution and reproducibility of a dedicated phantom for proton PBS daily quality assurance”. In: Zeitschrift für Medizinische Physik 28.4 (2018), pp. 310–317. issn: 0939-3889. https://www.ncbi.nlm.nih.gov/pubmed/29548595
3. Lynx factsheet: https://www.iba-dosimetry.com/product/lynx-pt/
4. Sphinx factsheet: https://www.iba-dosimetry.com/fileadmin/user_upload/products/03_proton_therapy/Sphinx/PT-QA_Sphinx_Flyer_EN_Rev.3_1017.pdf

IBA Zebra and Giraffe

The Zebra is a multilayer ionization chamber (MLIC) consisting of 180 parallel plate ionization chambers separated by 2 mm and the chambers have a diameter of 2.5 cm. Similarly the Giraffe is an MLIC consisting of 180 parallel plate chambers. The chambers in the Giraffe have a diameter of 12 cm so the Giraffe is a lot bigger than the Zebra.

Both these devices can be utilised for QA measurements of the depth-dose profile. The accuracy of the measurements is limited by the resolution of the detector. The range and SOBP width are measured with tolerances of 2 mm and 5 mm respectively.

Links to Papers

1. Sandeep Dhanesar et al. “Quality assurance of proton beams using a multilayer ionization chamber system”. In: Medical Physics 40.9 (2013), p. 092102. doi: 10.1118/1.4817481. https://www.ncbi.nlm.nih.gov/pubmed/24007171
2. Christian Baumer et al. “Evaluation of detectors for acquisition of pristine depth-dose curves in pencil beam scanning”. In: Journal of Applied Clinical Medical Physics 16.6 (2015). https://www.ncbi.nlm.nih.gov/pubmed/26699567
3. Zebra datasheet: https://www.iba-dosimetry.com/product/zebra/
4. Giraffe datasheet: https://www.iba-dosimetry.com/product/giraffe/

Dosimetry

Very insightful paper on different dosimetric techniques that can be utilised in ion beam therapy. Christian P Karger et al. “Dosimetry for ion beam radiotherapy”. In:Physics in Medicine & Biology 55.21 (2010), R193. https://www.ncbi.nlm.nih.gov/pubmed/20952816

Research Essay Clinical Procedures Summary

PSI

This method is for pencil beam scanning and takes 20 minutes per gantry. A combination of detectors are used as part of the QA procedure including an MLIC, two strip ionization chambers, two scintillating screens and a phantom block made of PMMA material containing 2 ionization chambers. The device can be easily rotated to different gantry angles.

Property measured	Component	Description	Precision
Dose Output	Ionization Chambers	The ionization chambers are contained in the PMMA phantom. First ionization chamber is positioned at centre of SOBP and measures the total dose absorbed. The second is placed in the 50 % fall off region.	1 %
Range	MLIC	The MLIC contains 128 parallel plate ionization chambers.	2 mm
Spot Position	Strip ionization chamber and Scintillator	Monitored visually by eye.	1 mm
Beam Width and Size	Strip ionization chamber and Scintillator	The beam size and shape is monitored visually by eye.

Reference

1. O Actis et al. “A comprehensive and efficient daily quality assurance for PBS proton therapy”. In: Physics in Medicine & Biology 62.5 (2017), p. 1661. https://www.ncbi.nlm.nih.gov/pubmed/28166055

M.D Anderson

Property measured	Component	Description	Precision
Dose Output		Output measured at the centre of the SOBP.	2 %
Distal Range	Faraday Cup	The Faraday Cup is placed in the nozzle and the range is defined to be where the distal dose is at 10 %.	10 %/ 3 mm
Spot Position	Matrixx	Matrixx is in a solid water phantom and also allows a 2D dose profile to be obtained.	2 %/ 2 mm

Reference

1. Bijan Arjomandy et al. “An overview of the comprehensive proton therapy machine quality assurance procedures implemented at The University of Texas M. D. Anderson Cancer Center Proton Therapy Center–Houston”. In: Medical Physics 36.6Part1 (2009), pp. 2269–2282. https://www.ncbi.nlm.nih.gov/pubmed/19610316

MedAustron

Similarly to PSI, MedAustron use a combination of commercially available detectors. The detectors used are the Lynx, Giraffe and an ionization chamber in solid water.

Property measured	Component	Description	Precision
Range	Giraffe	Depth dose profiles can be obtained from the Giraffe.	2 mm
Dose Output	Roose Chamber	The chamber is placed at a reference depth.	2 %
Spot Position	Lynx	Visually measured by eye.
Spot Size	Lynx	Visually measured by eye.	2 mm

Reference

1. Loıc Grevillot et al. “Implementation of dosimetry equipment and phantoms at the MedAustron light ion beam therapy facility”. In: Medical physics 45.1 (2018), pp. 352–369. https://www.ncbi.nlm.nih.gov/pubmed/29105791

UCLH

UCLH is set to treat its first patient in 2020 and is currently in the planningand commissioning phase prior to treatment. As a new centre, it is initially planned for the QA procedures at UCLH to follow and learn from existing methods in place at functioing centres. UCLH will have 4 treatment rooms and ideally, the daily QA process should take about 30 minutes per room. For pencil beam scanning treatment, the beam characteristics that should be checked daily are the beam range, output, spot position and spot size. Due to cost considerations and practicality, the Sun Nuclear QA3 device has initially been chosen to carry out the daily QA measurements. The key objective for daily QA at UCLH is to look for large deviations in the measured characteristics. The tolerances in the range, output and spot size have not been determined but they are likely to be relatively large compared to other centres. However, the tolerance on the spot position is planned to be much tighter due to the direct effect small deviations in the position have on the uniformity of the intensity distribution. The daily QA procedure will be similar in principle to other centres that use the Sun Nuclear QA3 device.

Tasks Completed

• Presentation to PBT group.
• Project progress report.
• Project outline.
• First plan for research essay.
• Prepare practise presentation to Agapi.
• Email Alison for details of Haakan Nystrom.
• Met with Alison and Andrew form UCLH
• Research Essay
• Added links to wiki.
• Summarised research essay in wiki.

Key Dates

Task	Deadline
Project Outline	03/12/2018
Project Progress Report	31/01/2019
Short Presentation	08/02/2019
Research Essay	31/03/2019
Thesis Deadline	22/08/2019
Final Presentation	02/09/2019

Second Meeting With Alison and Andrew (UCLH)

Firstly I updated Alison and Andrew on what progress I had made so far in my project. We briefly discussed some of the devices that I had come across as part of my research and the common beam characteristics that had been measured as part of daily QA.

From my Research

As before the characteristics measured as part of daily QA were:

• The beam energy.
• Output.
• Spot position (not so much uniformity as difficult to do).
• Spot size (and to an extent shape).

Summary of detectors:

• Sun Nuclear DQA3: The DQA3 is a diode and ion chamber device used to measure the beam output, range, symmetry, spot position and size. Very useful device.
• IBA Matrixx: Parallel plate chamber based device used to measure output, beam flatness and symmetry. Not as useful.
• IBA Zebra: Multilayer ionization chamber device. Not as useful.
• IBA Lynx: Scintillator based device used to measure spot size and position. Very useful device however limited due to cost.
• IBA Sphinx and Lynx: Combination of Lynx and Sphinx used to measure spot position, size and range. Very useful but again limited by cost.
• Gaffchromic film: Film based detector used to obtain dose profile.

Clinical Centres:

• PSI use a Lynx and Sphinx combination for daily QA.
• Med Austrom use a Giraffe, Lynx and chamber in solid water on a couch for daily QA.
• Both of these are very expensive approaches.

Discussion

Andrew and Alison confirmed that they planned to initially use the Sun Nuclear DQA3 device for their daily QA measurements. This decision was based on a compromise between cost and which measurements could be carried out. The IBA Lynx and Sphinx is the "gold standard" in daily QA systems however would cost roughly £500,000 which is not feasible. Conversely, the DQA3 device is about £8,000 and has the capability to perform the required measurements.

The output and range measurements are carried out using the central 5 ionization chambers and the spot position and size with the diodes. The range measured is a measure of range fluctuations rather, this is sufficient for daily QA. Different buildups are placed in front of the ionization chambers to allow for different range measurements. The diodes used are at a set depth and position and so get limited information from them but could be sufficient for daily QA. To measure the position the beam is directed at the innermost diode of the triplet and a half Gaussian is constructed by the detection from the triplet of diodes. This is combined with the half Gaussian obtained from the triplet on the opposite side and if the combined shape is roughly Gaussian then that acts as a confirmation of spot position. There are difficulties measuring the spot size as the beam will not be circular when measured, and, for example, could be elliptical in shape and there are therefore difficulties in obtaining the size directly (a scintillator would be better to get size).

The spots delivered are ideally Gaussian, and the spots should be separated such that the intensity is constant to roughly 2/3 %. The intensity cannot be measured with the DQA3. The intensity distribution is very sensitive to spot position deviations. Small changes in the spot position can greatly effect the uniformity of the intensity distribution. This increases the precision at which the position must be measured.

In order to activate the DQA3, the central axis chamber, used to measure the output, must be irradiated to trigger measurements. Can either irradiate the whole surface and perform measurements. Or quickly perform measurements on diodes after irradiation. One option is to obtain multiple data sets/ fields and then put together to reconstruct data.

Monthly QA was also discussed but they did not have an ideal set of measurements to take for this and is dependent on what they get out the daily QA. One option is to use the Zebra (multilayer ionization chamber).

From daily QA, the goal was to look for big deviations for the energy, beam profile and output, apart from with the spot spacing (position).

For me to potentially work on in the future:

• Look at putting different build up in front of the chambers to generate different depths. The buildups are made up of plastic WET material. In the patient the dose distribution is dominated by the scattering interactions rather than the beam energy.
• Look at doing a DQA3 prototype and maybe test it in Denmark.

Emails with Hakan (Skandionkliniken)

I emailed Hakan regarding the daily QA procedures at the PBT facilities at Skandionkliniken and Aarhus. We discussed essential daily QA checks and he confirmed that the:

• Beam energy (range)
• Spot positon
• Spot size
• Output
• Imaging systems

Something that I had left out from my research so far was how the daily QA of the imaging systems. Modern PBT is very image guided and so if the imaging geometry is out then the treatment will also be out. Furthermore, a geometrical offset of the imaging system is likely and happens all the time, and so needs to be incorporated into daily QA.

Hakan also confirmed what Alison and Andrew had alluded to about the beam energy (range). Although it is an important parameter, it is very unlikely to go wrong due to technical reasons (true for cyclotron systems). Therefore he suggested that the energy could even be left to be checked on a weekly or monthly basis.

A viable QA system should take no more than 30 mins per gantry. The QA at Aarhus takes about 15 mins and is dependent on the parameters and the number of irradiations.

Emails with Alessio (MedAustron)

Discussed daily QA procedures at MedAustron. Daily QA is split into three categories of measurement:

• Optics - spot position and size
• Depth dose profiles - range, energy and energy spread
• Dose

A combination of detectors are used for daily QA. A 2D scintillator (Lynx) is used for optics measurements, Giraffe for depth - dose profiles, and a ROOS chamber in plastic at a reference depth for the dose.

Tolerances for measurements are based on commissioning data, shared experience with other centres and current literature. The maximum current tolerances allowed are:

• 2 mm or 40 % for spot size
• 2 mm for range at reference depth of 80 % dose level
• 2 % in dose

Comments made about presentation

• Understand what range modulators do.
• Look into whether Zebra was initially designed for X-Rays.
• Zebra not often used for daily QA due to the very long initial set up time.
• Investigate film dosimetry e.g Gafchromic film.
• Research Lynx scintillator.
• Range calorimeter - need a moving absorber, e.g water tank, to measure energy and range.
• Know difference between scintillator screens and fibres - fibres use fibre optics and are not used due to needing something to read output from each fibre which is expensive.
• UCLH will have 4 treatment rooms.
• Put slide number on presentation.
• Expand on each daily QA:
  • Table of devices
  • Tolerances
  • How long it takes
• Collect standards and tolerances on energy, spot size and output.

Feedback and improvements on Research Essay

Feedback from Raffy:

• Include more figures to explain points, e.g for difference between cyclotrons and synchrotrons.
• Split paragraphs into smaller subsections to break up text, especially in introduction.
• Define terms before using them, especially medical terms such as gantry, gantry angle as well as terms like output.
• Use lists to clearly show variables as easier to find quickly.
• Include a short section on different detectors before introducing them, e.g what is an ionization chamber, distinguish roose, farmer, strip chambers. Also talk about gas used in chambers (air, argon etc)
• When describing things make sure to define them first, e.g water phantoms.
• Important to explain acronyms.
• Summary table with variables and tolerances would be useful.
• Label figures identifying key parts. Could do this on powerpoint/word and then adapt caption to say image adapted from.
• Include a map of centres so clear where talking about and to show distribution.
• Generally include pictures if can, particularly when talking about stepped wedge to make clear to reader.
• When writing in text about private communication make it explicit who you were talking to and also in references mention their role.
• Ask Saad to see his report.

Feedback from Matthew Wing:

• Last paragraph from intro was too long and could be split into smaller ones.
• Overview of what a PBT facility looks like showing beam line, gantry etc with pictures.
• Discussion and visualisation of how and where diagnostic systems fit into the overall scheme would have been useful and not just an isolated picture of a give detector.
• Could have a section in intro about physics of protons in matter and why curve is that shape - due to interactions.
• Abstract as a summary at beginning to summarise the whole point.
• Better quality images would be good.

Dissertation Plan

Structure

• Abstract
• Introduction:
  • Radiotherapy background
  • Physics behind PBT
  • Advantages of PBT
  • QA and why its needed
• Beam Characteristics - what's measured and why
  • Output
  • Range/ Energy
  • Spot position and size
  • Flatness and Symmetry
  • Further daily QA checks
• Detector types (and examples of commercially available ones and examples of clinical procedures with them)
  • Scintillator: Lynx, plastic scintillator
  • Ionization Chambers: Thimble, farmer, parallel plate, MLIC
  • Diodes
  • Film dosimetry
• Potential detector designs
  • Scintillator + phantom
  • Scintillator volume
  • Scintillator stack
  • Diodes + ion chambers
  • 2 strip chambers orientated perpendicular to each other.
• Future work
• Summary

Timeline

• Introduction, beam characteristics and detector types written by 21/07
• Some initial drawing designs to show Simon by 17/07

Potential Design Blueprint

• Some kind of diagram
• Components and their size and position relative to each other
• Justification
• What measurements it would be capable of and precision
• Limitations and considerations
• Similar designs investigated in literature.
• How could it be improved