Measurements of the Variation of RFKO Extracted Beam Current for Hitachi Proton Therapy and Carbon Therapy Synchrotrons and Implications for Particle Therapy

Measurements of the extracted beam current (BC) for a Hitachi Carbon Therapy Synchrotron and a Hitachi Compact Proton Therapy Synchrotron are reported for the nominal extracted beam current (BC0) of 10mMUmsec , which was chosen since it is the magnitude of the clinical BC0, and a sample rate of 5 μsec (Carbon) and 8 μsec (Proton). A Noise Power Spectrum (NPS) analysis identifies the source of variation to be beam or power supply related. The rise time in the BC has been modeled and its effect on beam delivery time simulations have been estimated. The impact of the variation in BC from BC0 is shown to cause potentially significant dosimetric uncertainties in treatment delivery for modern particle therapy accelerators using fast Scanning Magnet (SCM) if plans are not simply beam current moderated or robustly optimized. The variation in beam current is shown to be inconsequential for medical physics quality assurance and commissioning measurements using properly biased ion chambers.


Introduction
Pencil beam scanning (PBS) using discrete spot scanning (DSS) delivery was the method that was introduced to proton clinic use in 1995 at PSI [1].Pencil beam raster scanning irradiation is the common scanning mode for Carbon PBS in modern heavy ion therapy institutions [2] [3].Measurements of the irradiated dose in the transition between the spots called Extra Dose in Raster Scanning (EDR) was reported for HIMAC at NIRS/QST in 2007 [4].Hitachi's pencil beam raster scanning is available for their Compact Proton Synchrotron [5] and Carbon Synchrotron [6] and is called dose-driven continuous scanning (DDCS) [7].The EDR referred to by Inaniwa is called Move Dose for Hitachi Synchrotrons.Raster scanning requires an understanding of the extracted beam current over the relevant time domain of the spot delivery.Slow extraction is known to have a spill ripple and in the Japanese particle therapy systems in particular the RF-knockout (RFKO) beam extraction technique is used.
In addition to these developments there is also technological advancement and power supply improvements such that the scanning magnet scan speed is improved in the last decade.This fast scanning speed implies a very short time domain of the spot delivery and the spill ripple should be considered.Mayo Clinic Florida (MCF) will install the latest Hitachi particle therapy system in a few years' time.In light of this, we conducted the current study in collaboration with Hitachi engineers to measure the variation of the extracted beam current of Hitachi synchrotrons.

Methods
Measurements of the extracted beam currents were performed for a Hitachi Carbon Synchrotron and a Hitachi Compact Proton Synchrotron.The Hitachi Carbon Synchrotron is 56.8 meters in circumference and comprised of 12 dipoles.The Hitachi Compact Proton Synchrotron comprised of either six or four dipoles.The measurements reported here are for the Hitachi four dipole Compact Proton Synchrotron which is 18.0 meters in circumference.The Hitachi Synchrotrons use the RFKO for extraction and a feedback from the extracted beam current measured in the Dose Monitor (DM) to adjust and stabilize the BC.When the request for extraction is received by a control point the extraction is initiated.A new control point can be either: (1) at the beginning of a spill; (2) at the beginning a new energy layer when utilizing multi energy extraction (MEE) or; (3) a Breakspot.As described in [7] a Breakspot, by way of analogy to the conventional DSS delivery, refers to a spot that, in a given layer, the beam extraction is halted before moving to a new spot position.
The DM is in the nozzle of the Carbon or Proton port and is just slightly downstream of the vacuum exit window.It is vented to air and operated at several kV bias voltage.The DM current is converted to pulses via a current to frequency converter and therefore the number of pulses are proportional to the charge collected in the DM chamber.The charge per pulse is approximately 0.5 pC and this is then converted to a Monitor Unit (MU) via an adjustable calibration constant.One pulse is approximately 10 −6 MU or 10 −3 mMU.Further details of the Hitachi DM are proprietary.For this study the beam was being continuously extracted from the synchrotron and the number of pulses (ie the DM charge and therefore the MUs) is counted at regular intervals.For the Carbon Synchrotron measurement it is a period of every 5 µsec (or 200 x 10 3 times per second) and for the Compact Proton Synchrotron measurement it is a period every 8 µsec (or 125 x 10 3 times per second).To be clinically meaningful the measurement period needs to be shorter than, or comparable to, the typical times between spots.As described in [4] and [7] the actual move time between delivered spots is: T M OV E = (SpotSpacing)/(SCM Speed) where SCM Speed is the speed of the spot movement at isocenter and the M U M OV E (also called Move Dose) is defined as: M U M OV E = BC × T M OV E .Some typical scanning magnets are on the order of 10 mm/msec at isocenter for the clinical range of particle energy.For these typical scanning magnets and assuming a typical spot spacing of 5 mm for particle plans, T M OV E can be as small as 0.5 msec.Some modern systems such as Mayo Clinic Florida the SCMSpeed for 70 MeV protons is approximately 650 mm/msec.

BC as a function of time during a spill averaged over different time domains for the Hitachi
Compact Proton Synchrotron for a BC0 of 10 mMU/msec is shown in Fig. 1.Fig. 2 shows the NPS of the finest resolution of Fig. 1 with labeling of the sources.The large amplitude low frequency labeled as Beam Origin is due to the RFKO and likely other consequences of extraction.The small amplitude high frequency is due to the noise in the magnet power supplies.The NPS for Carbon was observed to be similar to the Proton.In Fig. 3 is shown the scope trace of several spills of 6 seconds length for 430 MeV/u Carbon of BC0 of 10 mMU/msec showing that the maximum amplitude of BC variation was observed to increase over the course of the spill.These measurements were repeated for different BC0 and the Peak Dose Rate BC M AX versus BC0 is shown in Fig. 4.

Ramp Up Dosimetry
BC in the ramp up region can be described by a simple linear increase: where t ramp is the ramp up time.t ramp for Carbon was observed to be approximately 10 msec in Fig. 5. BC at the start of the spill for the Compact Proton Synchrotron was not recorded however it was observed that the typical t ramp was 3 msec since Proton RFKO approaches stability more quickly than Carbon RFKO.The extracted ramp up dose following a new control point can be estimated as: This means if the spot after a new control point is less than approximately 50 mMU then the spot delivery will be in a region of BC that is still ramping up to the nominal value of BC0 = 10 mM U msec .Moreover if the Carbon beam delivery always uses breakspots which are less than the ramp up dose then the 100% breakspot beam delivery will have an effective BC less than half of BC0.This will likely not have a significant impact on typical patient dosimetry except that it affects beam delivery time simulations for treatment plans that use 100% breakspots and it would also impact patient dosimetry simulations for moving targets with the commonly known interplay effects.

Move Dose Dosimetry
A(t) and B(t) are time dependent indicating changing peak dose rate and baseline as observed in Fig. 3.The frequency f is chosen to model the dominant fluctuation due to the RFKO which is a complex of frequencies on the order of kHz.This function is shown in Fig. 6 (c) where f = 1 kHz.We propose that the BC can be described by a simple time dependent function comprised of a linear ramp and a trigonometric variation.This model could be used to study how the variation in BC and move dose affects dosimetry and treatment delivery time for actual patient plans.Such a study for actual patient plans is beyond the scope of this work.

Quality Assurance Measurements
There is potentially an implication on quality assurance and commissioning measurements taken during the course of a spill due to the variation in collection efficiency of ion chambers with beam intensity.It is well known that the collection efficiency for ion chambers such as Bragg peak chambers, Markus chambers and Farmer type chambers changes slowly with BC so long as the bias voltage is set such that the collection efficiency at BC0 is nearly unity, for example 99.5%.Boag's equation [8] demonstrates that variation in collection efficiency should not be a concern for such well biased ion chambers.In such a case then the doubling or even tripling of the beam current should be still acceptable but nevertheless tested and validated.

Figure 1 .
Figure 1.140 MeV Proton Beam Current measurement for a target beam current of 10 mMU/msec integrated over different time domains.

Figure 2 .
Figure 2. FFT of Proton Beam Current showing contributions to the signal from Power Supplies and from Beam Variations.

Figure 4 .
Figure 4. Maximum observed beam intensity observed on scope trave for Carbon 430 MeV/u and different target beam intensities.

Figure 5 .
Figure 5. Ramp up of the Carbon 430 MeV/u beam.

Fig. 6
is a zoom in showing just one msec of BC for BC0 = 10 mM U msec for (a) Hitachi 430 MeV/u Carbon Synchrotron and (b) Hitachi 140 MeV Compact Proton Synchrotron.Following the ramp up region the beam current can be reasonably described by a function like