Electrostatic shielding in ultra-high dose rate electron radiotherapy

Abstract

A medical linear accelerator configured to operate in radiotherapy may have an electron gun for emitting an electron beam. An accelerator waveguide accelerates the electron beam. A linear accelerator head defines an inner beam passage for directing the electron beam. A beam current transformer in the linear accelerator head is for monitoring the electron beam An electrostatic shield is provided in the inner beam passage, the electrostatic shield being positioned downstream of the beam current transformer for shielding the beam current transformer.

Description

ELECTROSTATIC SHIELDING IN ULTRA-HIGH DOSE RATE ELECTRON RADIOTHERAPYCROSS-REFERENCE TO RELATED APPLICATION

[0001] The present application claims the benefit of United States Patent Application No. 63 / 570,939, filed on May 28, 2024, the content of which is incorporated herein by reference.TECHNICAL FIELD

[0002] The application relates to medical linear accelerator for radiotherapy, and more particularly to ultra-high dose rate (UHDR) electron radiotherapy.BACKGROUND

[0003] Beam current transformers (BCT) are promising detectors for real-time beam monitoring in ultra-high dose rate (UHDR) electron radiotherapy. However, BCT signals have exhibited a significant sensitivity to changes in source-to-surface distance (SSD), field size, and phantom material which may be attributed to the fluctuating levels of electrons backscattered within the BCT, among other phenomena.

[0004] Radiotherapy has long been a cornerstone in the treatment of various malignancies, offering a non-invasive approach to target and eradicate tumor cells. Over the years, advancements in technology and techniques have sought to maximize the therapeutic ratio by enhancing tumor control while minimizing damage to surrounding normal tissues. One of the most recent and potentially transformative advancements in this direction is the discovery of the FLASH effect. The FLASH effect refers to the observation that UHDR irradiation, delivered at rates above 40 Gy / s, can achieve equivalent tumor control while substantially reducing normal tissue toxicity. This phenomenon has been observed across various pre-clinical models, including invertebrates, rodents, and larger mammals. The underlying mechanisms for the FLASH effect are still under investigation, but hypotheses include differential oxygen consumption, enhanced DNA repair in normal tissues and alterations in the immune response. While the FLASH effect has been observed for photons, electrons and protons, most research evidence stems from MeV electron beams. This is due to the wide range of machines capable of generating UHDR electron beams including dedicated accelerators, converted conventional linear accelerators and intra operative radiotherapy machines. Electron UHDR radiotherapy consequently represents thecurrent reference for clinical transfer of FLASH radiotherapy in pre-clinical and clinical settings.

[0005] A significant challenge in the translation of UHDR radiotherapy is the limited ability to accurately measure key irradiation parameters, such as dose and dose rate, in UHDR radiation beams using standard radiation detectors. Similarly, real-time monitoring and control of a UHDR beam output present a challenge as monitor ionization chambers used in conventional linac heads fall short in UHDR beamlines due to saturation and ion recombination effects. Novel strategies and detectors have been investigated to enable real-time monitoring of UHDR beams, including Cherenkov imaging, plastic and inorganic scintillators, probe calorimeters, and BCTs. BCTs provide a real-time monitoring solution without causing perturbations or experiencing saturation effects. BCTs consist of a conducting winding wrapped around a toroidal ferromagnetic core, where a voltage proportional to the beam current in the central axis of the toroid is generated through electromagnetic induction. One notable advantage of BCTs over transmission chambers in UHDR beam monitoring is their capability to verify the beam’s parameters, such as the number of pulses, pulse width, and pulse repetition frequency, while potentially being able to correlate the measured current or charge with the absorbed dose at a specific point downstream of the BCT. For this reason, electron UHDR beam monitoring using BCT is being adopted by several groups.

[0006] Despite the potential of BCT for beam current monitoring, the high sensitivity of BCTs to variable irradiation conditions as a potential limitation. Specifically, factors such as SSD, phantom material, and collimator size have been observed to influence the BCT readings by up to 12%, potentially compromising their usability for beam output monitoring. To date, backscattered radiation has been suggested as the main contributor to these discrepancies, as backscattered electrons traveling back into the toroid would reduce the net current measured by the BCT. However, other factors may contribute to the sensitivity of BCT signal against variations in irradiation conditions.SUMMARY

[0007] In a first aspect, there is provided a medical linear accelerator configured to operate in radiotherapy, comprising: an electron gun for emitting an electron beam; an accelerator waveguide for accelerating the electron beam; a linear accelerator head defining an inner beam passage for directing the electron beam; a beam currenttransformer in the linear accelerator head for monitoring the electron beam; and at least one electrostatic shield in the inner beam passage, the electrostatic shield being positioned downstream of the beam current transformer for shielding the beam current transformer.

[0008] Further in accordance with the first aspect, for instance, the linear accelerator head has a plurality of collimators.

[0009] Still further in accordance with the first aspect, for instance, the beam current transformer is in one of the collimators.

[0010] Still further in accordance with the first aspect, for instance, the linear accelerator head includes an applicator.

[0011] Still further in accordance with the first aspect, for instance, the beam current transformer surrounds or is placed within the applicator.

[0012] Still further in accordance with the first aspect, for instance, the applicator has a downstream portion and an upstream portion, the downstream portion and the upstream portion interconnected to one another.

[0013] Still further in accordance with the first aspect, for instance, the downstream portion and the upstream portion define a space to receive the beam current transformer therein.

[0014] Still further in accordance with the first aspect, for instance, the at least one electrostatic shield is in the applicator.

[0015] Still further in accordance with the first aspect, for instance, the at least one electrostatic shield is laid flat against at least one axial face of the beam current transformer.

[0016] Still further in accordance with the first aspect, for instance, the at least one electrostatic shield is a sheet.

[0017] Still further in accordance with the first aspect, for instance, the at least one electrostatic shield is a sheet of conductive polymer.

[0018] Still further in accordance with the first aspect, for instance, the conductive polymer is one of polyimide, polyester and polyethylene.

[0019] Still further in accordance with the first aspect, for instance, the at least one electrostatic shield is a metallic sheet.

[0020] Still further in accordance with the first aspect, for instance, the at least one electrostatic shield is a graphene sheet.

[0021] Still further in accordance with the first aspect, for instance, the electrostatic shield lies in a plane across the inner beam passage.

[0022] Still further in accordance with the first aspect, for instance, a central axis of the inner beam passage is normal to the plane.

[0023] Still further in accordance with the first aspect, for instance, the electrostatic shield is grounded to earth.

[0024] Still further in accordance with the first aspect, for instance, two of the electrostatic shield are provided, one of the electrostatic shield being downstream of the beam current transformer, and another one of the electrostatic shield being upstream of the beam current transformer.

[0025] Still further in accordance with the first aspect, for instance, a bending magnet is configured to direct the electron beam in the inner beam passage.

[0026] In accordance with a second aspect of the present disclosure, there is provided a method for operating a medical linear accelerator in radiotherapy comprising: within the medical linear accelerator, emitting an electron beam, accelerating the electron beam, directing the electron beam via an inner beam passage, monitoring the electron beam in the inner beam passage by a beam current transformer, outletting the electron beam from the medical linear accelerator, and shielding the beam current transformer at least from backscatter.

[0027] Further in accordance with the second aspect, for instance, directing the electron beam includes collimating the electron beam.

[0028] Still further in accordance with the second aspect, for instance, collimating the electron beam includes collimating the electron beam in at least two collimators.

[0029] Still further in accordance with the second aspect, for instance, directing the electron beam includes bending the electron beam to the inner beam passage.

[0030] Still further in accordance with the second aspect, for instance, the emitting of the electron beam may be as a function of a value of the monitoring of the electron beam by the beam current transformer.

[0031] Still further in accordance with the first aspect, for instance, the beam current transformer may be shielded from energy accumulated in the medical linear accelerator upstream of the beam current transformer.DESCRIPTION OF THE DRAWINGS

[0032] Reference is now made to the accompanying figures in which:

[0033] Figs. 1 A and 1B are schematic views of medical linear accelerators operating at ultra-high dose rates and using electrostatic shielding in accordance with the present disclosure;

[0034] Fig. 2A is an enlarged view illustrating an exemplary position of electrostatic shields in an applicator of the medical linear accelerator, in accordance with a variant;

[0035] Fig. 2B is an enlarged view illustrating an exemplary position of a single electrostatic shield in an applicator of the medical linear accelerator, in accordance with another variant;

[0036] Figs. 3A and 3B are schematic views of other types of medical linear accelerators operating at ultra-high dose rates and using electrostatic shielding in accordance with the present disclosure;

[0037] Fig. 4 is a Monte Carlo calculated fraction of incident electrons backscattered (a) on a water phantom as a function of energy and (b) on different materials for energies of 0.75 and 9 MeV;

[0038] Fig. 5 is a mean integral BCT signal per pulse as a function of SSD for beam energies of 6 MeV and 9 MeV, using the test applicator without electrostatic shielding. The black bars represent one standard deviation for the repeated measurements;

[0039] Fig. 6 is a mean integral BCT signal relative to the 6 cm field size for beam energies of 6 and 9 MeV using the clinical applicator without electrostatic shielding. The black bars represent one standard deviation for the repeated measurements;

[0040] Fig. 7 is a mean BCT signal per pulse for various source-to-surface distance (SSD) for water and Solid Water® phantoms using a 9 MeV UHDR beam without electrostatic shielding. Each plot also reports the ratio of the mean integrated signal in Solid Water® to the one in water;

[0041] Fig. 8 is a mean BCT signal per pulse for the irradiation of (a) Solid Water® and (b) water using a 9 MeV UHDR beam without electrostatic shielding and a source-to-surface distance (SSD) of 30 cm. The integral BCT signal for scenarios (a) and (b) is presented in (c), where the black error bars show one standard deviation for the repeated measurements;

[0042] Fig. 9 is a mean BCT signal per pulse for the irradiation of (a) Solid Water® grounded and ungrounded and (b) Solid Water® with a Roos cham-ber either connected or unconnected to a powered electrometer using a 9 MeV UHDR beam without electrostatic shielding. The bias applied to the chamber did not impact the BCT signal (not shown);

[0043] Fig. 10 is a mean BCT signal per pulse for irradiations of a Solid Water® phantom at various SSD using a Faraday shield (a) grounded and (b) ungrounded. The mean integral signal for scenarios (a) and (b) is presented in (c), where the black error bars show one standard deviation for the repeated measurements;

[0044] Fig. 11 is a mean BCT signal per pulse in water and Solid Water® for three irradiations of five pulses each at (a) 6 MeV and (b) 9 MeV when using a grounded Faraday shield. The mean integral BCT signal per pulse for these irradiations is shown in (c), with the black error bars representing one standard deviation for the repeated measurements; and

[0045] Fig. 12 is a mean integral BCT signal per pulse relative to the 6 cm field size with a grounded Faraday shield using the clinical applicator with a Faraday shield either grounded or ungrounded. The black error bars show one standard deviation for the repeated measurements.DETAILED DESCRIPTION

[0046] Referring to the drawings and more particularly to Fig. 1A, a medical linear accelerator in accordance with the present disclosure is generally shown as 10. The medical linear accelerator 10 may be referred to as a linear accelerator, as a linac, and is used to irradiate a lesion in the treatment of tumours. The medical linear accelerator 10 may for example be a Mobetron medical linear accelerator by IntraOp Medical Corporation. The medical linear accelerator 10 is shown schematically with some of its major components being illustrated, as the head of the medical linear accelerator 10 is the focus in Fig. 1A. However, other components of the medical linear accelerator 10 may not be illustrated for simplicity of the figures, such as the structural components supporting the operational components of the medical linearaccelerator 10. Moreover, additional components omitted from the figures may include the electronics, sensors, controller components, etc.

[0047] The medical linear accelerator 10 may include an electron gun 11, that may include a cathode. The cathode may be heated or powered during operation, via a controller 10A of the medical linear accelerator 10. An accelerating waveguide 12 may be aligned with the electron gun 11 , and located downstream of the electron gun 11. In the accelerating waveguide 12, low-energy electrons emitted by the heated cathode in the electron gun 11 are accelerated to given energy levels, such as in a beam of electrons, or electron beam. For example, the energy levels may be in the megavoltage range, typically 6-18 MeV, though these values are merely given as an example.

[0048] One or more scattering foils 13 may be located in and / or at an end of the accelerating waveguide 12. For example, Fig. 1A shows the scattering foil 13 at an exit window 12A of the accelerating waveguide 12. Once exiting the accelerating waveguide 12, the electron beam passes through the one or more scattering foils 13. The scattering foil(s) 13 may modify the beam profile. For example, the beam may go from a pencil beam to a diverging and relatively uniform broad beam.

[0049] A linear accelerator head (a.k.a., linac head) may be downstream of the accelerating waveguide 12, and is provided to guide or direct the electron beam. In the linear accelerator head, one or more collimators, referred to concurrently as collimators 14, may be located downstream of the accelerating waveguide 12. In Fig. 1 A, three collimators 14 are shown, serially arranged relative to one another. The collimators 14 are provided to adapt the shape of the beam to a targeted lesion. The series of collimators 14 may for instance include a primary collimator 14A, a secondary collimator 14B and a tertiary collimator 14C. In the illustrated embodiment, the tertiary collimator 14C may be the one held closest to the irradiated surface, by an applicator 15, but this is only an example. The applicator 15 may for example be used to place one of the collimators 14 directly upstream of the targeted lesion in order to optimize the conformality of the radiation dose. Other arrangements of the electron gun 11 , accelerating waveguide 12, scattering foil(s) 13 or like scattering device, collimator(s) 14, and applicator 15 may be used. For example, Fig. 3A shows an alternative arrangement in which such components are not in an in-line arrangement.

[0050] During use of the medical linear accelerator 10, a conventional dose rate setting may be selected. According to such setting, the beam intensity may bemonitored in real time by a monitor chamber 16, in the linear accelerator head. The monitor chamber 16 may measure the charge released by the beam therethrough, and may correlate it to the dose delivered to a point downstream of the linac head. A feedback system is part of the medical linear accelerator 10 and is used to allow the medical linear accelerator 10 to interrupt the beam once a prescribed dose has been delivered by the medical linear accelerator 10. In an ultra-high dose rate (UHDR) setting, the monitor chamber 16 may experience saturation effects, making it less accurate in real-time beam monitoring and dose reporting. Therefore, a beam current transformer 17 (BCT) is used for monitoring and dose reporting in the linear accelerator head. The BCT 17 may consist of a conducting winding wrapped around a toroidal ferromagnetic core, where a voltage proportional to the beam current in a central axis X of the BCT 17 (e.g., toroid in shape) is generated through electromagnetic induction. This is one possible construction of the BCT 17. The BCT 17 may be located at different positions in the medical linear accelerator 10. For example, in Fig. 1A, the BCT 17 is located on the applicator 15, between collimators 14B and 14C. In Fig. 1 B, the BCT 17 is positioned in the monitoring chamber 16, and / or in the collimator 14A. Other locations are possible. The BCT 17 may be used as an alternative to the monitoring chamber 16, in any of the embodiments herein, the monitoring chamber 16 optionally being present in any of the medical linear accelerators 10 described herein.

[0051] In order to shield the BCT 17, an electrostatic shield(s) 20 may be provided in the medical linear accelerator 10. The electrostatic shield 20 may be referred to as a Faraday shield, an electrostatic cage, among other names. The electrostatic shield 20 may extend across the inner beam passage of the medical linear accelerator 10, such that the electron beam must pass through the shield 20 when emitted by the medical linear accelerator 10. For example, the electrostatic shield(s) 20 lie(s) in a plane to which a central axis X of the inner beam passage of the medical linear accelerator 10 is normal, at the collimator(s) 14, as best seen in Fig. 2A and Fig. 2B. Other configurations and orientations may be used, with the shield 20 optionally positioned to redirect the electron beam. The shield 20 may be grounded, in any appropriate way, such as by a wire extending from the shield 20 to a ground, such as a Faraday cage. The shield 20 could also be kept at a constant potential using appropriate means. The shield 20 may have different forms. For example, the shield 20 may be a sheet, a foil, or like membrane, but could have other configurations (e.g., mesh). Different conductive materials may be used, such as a metal (e.g. aluminum),graphene, or conductive polymers. Some of the conductive polymers that may be used include conductive polyimide (e.g., Kapton®), conductive polyethylene, conductive polyester film (e.g., Mylar®). The characteristics of the shield(s) 20, such as thickness, material (e.g., metal such as aluminum, polymerwith coating, steel, etc), etc., are selected for the electron beam to pass through the shield 20 when irradiating a lesion, but to block any backscatter from reaching the BCT 17 and / or electric fields from transient charging of surrounding materials under UHDR electron irradiation. Suitable thicknesses as a function of material may be as follows, as examples only: aluminum (e.g., 0.025 mm), conductive polyimide (e.g., 0.025 mm), graphene (e.g., 0.025 mm), conductive polyethylene (e.g., 0.125 mm). The thicknesses may be greater or less than these values, again to shield the BCT 17 while allowing the electron beam to pass through the shielded BCT 17. Depending on the configuration, the shield 20 may be said to be upstream of the BCT 17 (as shown as 20’ in Fig. 2A) and / or downstream of the BCT 17 (as shown as 20” in Fig. 2B or as in Fig. 2B), relative to an emission of the electron beam. In Fig. 2A, there are two shields 20, i.e., one upstream and one downstream of the BCT 17. It may therefore be said that the BCT 17 is fully encapsulated, entirely encapsulated, or sandwiched between shields 20. The shield(s) 20 isolate(s) the BCT 17 from electric fields which may otherwise affect the readings from the BCT 17, whether as backscatter and / or as accumulated in the head of the linear accelerator 10. Moreover, even though the shield 20 is downstream of the BCT 17 in Fig. 2B in a single-shield configuration, it is possible to have a single shield 20 upstream of the BCT 17.

[0052] Figs. 3A and 3B show another type of medical linear accelerator 10, that may also use the shield 20. The medical linear accelerator 10 of Figs. 3A and 3B differ from the medical linear accelerator 10 in that the components are not arranged in an in-line configuration, as a bending magnet 18 may allow the medical linear accelerator 10 of Figs. 3A and 3B to have a L-shape. The medical linear accelerator 10 of Figs. 3A and 3B may for example be a T rueBeam medical linear accelerator by Varian Medical Systems. The bending magnet(s) 18 is(are) examples of additional components that may be present in the medical linear accelerator 10.

[0053] In order to shield the BCT 17, an electrostatic shield 20 may be provided in the medical linear accelerator 10 of Figs. 3A and 3B. The electrostatic shield 20 may be referred to as a Faraday shield, an electrostatic cage, a Faraday cage, among other names. Again, the electrostatic shield 20 may extend across the inner beam passageof the medical linear accelerator 10, such that the electron beam must pass through the shield 20 when emitted by the medical linear accelerator 10. In the medical linear accelerator 10 of Figs. 3A and 3B, the shield(s) 20 may be upstream and / or downstream of the BCT 17, in arrangements similar to that of Figs. 2A and 2B. The electrostatic shield 20 may lie in a plane to which the central axis X of the inner beam passage of the medical linear accelerator 10 is normal, at the collimator(s) 14, again with reference to Fig. 2A or Fig. 2B. Other configurations and orientations may be used, with the shield 20 optionally to redirect the electron beam. The shield 20 may be grounded, in any appropriate way, such as by a wire extending from the shield 20 to a ground. The shield 20 could also be kept at a constant potential through appropriate means. The electron beam passes through the shield 20 when irradiating a lesion, but blocks any electric fields from transient charging of surrounding materials under UHDR electron irradiation. In the medical linear accelerator 10, the shield 20 may be positioned downstream and / or upstream of the BCT 17, relative to an emission of the electron beam. The shield 20 isolates the BCT 17 from the electric field and like backscatter, and / or energy accumulated in the head of the linear accelerator 10, which may otherwise affect the readings from the BCT 17.

[0054] Referring to Fig. 2A, the configuration of the applicator 15 is shown relative to the BCT 17 and to the shields 20. The applicator 15 may have a pair of members, one of which is the downstream portion 15A and the other being the upstream portion 15B. The downstream portion 15A may define a cylindrical passage centered on axis X. Other shapes are possible. A clip or like retaining formation may be at a bottom of the downstream portion 15A, such that a collimator may be connected to an end of the applicator 15 in the manner shown in Figs. 1 A and 1 B. The upstream portion 15B may be connected to the downstream portion 15A in any appropriate way, with lugs shown in Fig. 2A at the periphery of the downstream portion 15A and upstream portion 15B for fasteners to secure one to the other, as one possible way. The upstream portion 15B also has a retaining formation at its top end for connection to adjacent components of the linear accelerator 10, and a passage, such as a cylindrical passage centered on axis X. A space (e.g. , an annular space or gap) is concurrently defined at or adjacent to a junction between the downstream portion 15A and the upstream portion 15B so as to receive the BCT 17 therein, along with the shield(s) 20. The BCT 17 may thus be centered relative to axis X. Moreover, a vector of the axis X may be said to be normal to a plane of the shield(s) 20, but this is optional, as other orientations are considered. It may be said that the BCT 17 and the shield(s) 20 areentirely located between a plane including an upstream most end of the applicator 15, and a downstream most end of the applicator 15, in an embodiment. In the illustrated embodiment, and in other embodiments, it may be said that the shield(s) 20 is(are) laid against an axial face(s) of the BCT 17. In Fig. 2A, the shields 20 are laid flat against the upstream and downstream axial faces of the BCT 17. This is optional as a spacer could be present.

[0055] In order to assess the effect of the electrostatic shield 20 on the medical linear accelerator 10, Monte Carlo simulations and experimental measurements were conducted with a UHDR-capable intra-operative electron linear accelerator to analyze the impact of backscattered electrons on BCT signal, and those of the shield 20. The potential influence of charge accumulation in media as a mechanism affecting BCT signal perturbation was further investigated by examining the effects of phantom conductivity and electrical grounding. The simulations are merely given as an illustration of the effect of electrostatic shielding, with the simulations conducted under a set of particular conditions. Other approaches to quantify or qualify the efficacy of the electrostatic shielding could be used, such as by performing a comparison of the BCT readings with measurements using an alanine dosimeter, as an example.

[0056] The medical linear accelerator 10 of Figs. 1 Ato 3B, or any other medical linear accelerator may be operated in radiotherapy (e.g., UHDR) according to a method or process that may include at least some of the following steps and / or actions: within the medical linear accelerator, emitting an electron beam, accelerating the electron beam, directing the electron beam via an inner beam passage, monitoring the electron beam in the inner beam passage by a beam current transformer, outletting the electron beam from the medical linear accelerator, and / or shielding the beam current transformer at least from backscatter. Directing the electron beam may include collimating the electron beam. Collimating the electron beam may include collimating the electron beam in two or more collimators, or in two or more collimating steps or stages. Directing the electron beam may include bending the electron beam to the inner beam passage. The emitting of the electron beam may be adjusted as a function of a value of the monitoring of the electron beam by the beam current transformer. The beam current transformer may also be shielded from energy accumulated in the medical linear accelerator upstream of the beam current transformer.

[0057] Monte Carlo simulations indicated that the fraction of electrons backscattered in water and on the collimator plastic at 6 and 9 MeV is lower than 1%, suggestingthat backscattered electrons alone cannot account for the observed BCT signal variations. However, experimental measurements confirmed previous findings of BCT response variation up to 15% for different field diameters. A significant impact of phantom type on BCT response was also observed, with variations in BCT signal as high as 14.1% when comparing measurements in water and Solid Water®. The introduction of an electrostatic shield 20 to medical linear accelerators 10 during testing mitigated the dependencies of BCT signal on SSD, field size, and phantom material.

[0058] Variations in BCT signal as a function of SSD, field size, and phantom material are likely driven by an electric field originating in dielectric materials exposed to the UHDR electron beam. The electrostatic shield 20 may effectively prevent these electric fields from affecting BCT signal, enabling reliable BCT-based electron UHDR beam monitoring.

[0059] For experimental measurements, the medical linear accelerator 10 used was a Mobetron (IntraOp, CA), a mobile linear accelerator designed for intraoperative radiation therapy with FLASH capability through its research console. The Mobetron medical linear accelerator 10 was operated to produce 6 MeV and 9 MeV UHDR electron beams. With this version of the console, the Mobetron can generate UHDR beams of pulse widths ranging from 1.0 ps to 3.8 s at a pulse rate frequency (PRF) between 5 Hz and 90 Hz. As the monitor chamber 16 was not reliable for beam monitoring in UHDR, the BCT 17 was used to enable the prospective determination of the number of pulses for the solid-state modulator and electron gun. During the irradiation, the control system oversees the synchronization of each pulse to guarantee consistency across various pulse widths, while also logging each pulse administered. Reproducibility of the beam output in this setting was shown to be within 1% for both energies. The BCT 17 in tests was model ACCT-S-082-H from Bergoz, France, with its own differential amplifier. The exemplary BCT 17 has a rise time of 108 ns, a bandwidth of 3.075 MHz, an inner diameter of 4.1 cm and a signal drop of - 0.66% / ms. The BCT 17 positioned at the exit of the primary collimator and held in place using custom 3D-printed applicators. These applicators, described in more detail in the next section, were made with polylactic acid (PLA) in a tube-like geometry with a 1 cm wall thickness. At the distal end of the applicator, the electron beam can be shaped by a 4 cm thick collimator made of Delrin, with apertures ranging from 2.5 cm to 6 cm diameter. This setup was selected to ensure a reproducible yet removableinstallation of the BCT on the medical linear accelerator 10. A digital oscilloscope (DT5751 , 147 CaEN S.p.A., It) was used to measure the voltage out of the differential amplifier provided with the BCT 17, from which pulses were automatically detected using a fixed threshold of 0.05 V to trigger pulse recording. The signal from 2.75 ps before the trigger to 10 ps after the trigger was recorded for each pulse, for a total recording length of 12.75 ps per pulse. The readings were then processed using a custom Matlab script to derive the total BCT signal per pulse, defined as the integral of the BCT signal minus the average signal during the first 2 ps (i.e. the baseline signal) throughout the whole pulse recording.

[0060] During testing, two custom 3D-printed applicators were used. The applicators were designed to hold and center the BCT around the electron beam at a distance of 2.5 cm out of the linac head (i.e., from the exit of the accelerating waveguide 12, where the beam is collimated and monitored). The test applicator had a total length of 6.32 cm and no collimator holder in order to enable SSD measurements as short as 25 cm. A clinical applicator was optimized to deliver a dose of 3 Gy per pulse across field sizes of 2 cm to 4 cm, with a length of 15.0 cm (SSD = 33.3 cm), and was used to assess the impact of collimator size on BCT signal. UHDR electron beams of 6 and 9 MeV were used in testing, using a pulse length of 2.4 ps and a PRF of 60 Hz for all irradiations. Three irradiations of five pulses were delivered for each conditions listed below.

[0061] The effect of SSD on BCT signal was assessed by performing measurements with the BCT 17 held by the test applicator at SSDs between 25 cm and 70 cm (BCT to surface distance of 6.5 cm to 51.5 cm), using slabs of Solid Water® (Gammex®, Middleton, Wl) to define the surface. Measurements were done at 6 MeV and 9 MeV.

[0062] The effect of field size on BCT signal was evaluated by performing irradiations in air using the clinical applicator at 6 MeV and 9 MeV, using Delrin collimators with apertures of 2.5 cm, 4 cm, and 6 cm to shape the beam.

[0063] The influence of the phantom material on the BCT response was assessed by comparing the signal captured when irradiating Solid Water® and liquid water in a 68.0 cm x 40.7 cm x 35.0 cm water tank. Measurements were done at SSD ranging between 30 cm and 50 cm, using the test applicator to hold the BCT 17. Finally, BCT measurements were also acquired for the irradiation of a 5 cm Solid Water®) slab grounded through an aluminium foil and a conductive wire connected to a grounded Faraday cage protecting the electronics of the Linac, as shown in Figure 2. Similarly,a Solid Water® phantom with a plane-parallel Roos chamber (PTW, Freiburg, Germany) inserted at 1.6 cm depth was irradiated with the chamber both connected and unconnected to a powered Cardinal Health electrometer using biases of -300 183 V, 0 V and 300 V. BCT signal was recorded for all conditions, using an SSD of 25 cm and a beam of 9 MeV for both the irradiation of the grounded Solid Water® and that of the Solid Water® with a Roos chamber.

[0064] Monte Carlo simulations were performed using the EGSnrc framework with the usercode backscatter_clrp. This user-code is specifically optimized for calculating the backscatter coefficient resulting from a monoenergetic pencil beam of charged particles incident at a defined angle upon a target material. The backscatter coefficient q, in this context, refers to the probability of an incident particle scattering back into the hemisphere above the designated target. The purpose of utilizing this user code was to validate and investigate the behavior of electrons at various energy levels when interacting with different target materials. The coefficient q represents the worst-case scenario of the perturbation backscattered electrons could induce on the BCT signal for each configuration, as scatter in air as well as the limited aperture of the BCT would reduce the ratio of backscattered to primary electrons traveling back through the toroid.

[0065] Electron beams with energies ranging from 50 keV to 9 MeV were employed, oriented perpendicularly to the target surface. The number of histories for each simulation was chosen to achieve a statistical uncertainty of less than 0.5%, resulting in a range of 1 ,000,000 to 50,000,000 histories depending on the energy. The target thickness for all simulations was consistently set to 5 cm, and cross-section data used were those provided with EGSnrc’s version 4 installation. Default Monte Carlo transport parameters were utilized, incorporating all low-energy physics capabilities available within EGSnrc. Table 1 reports relevant simulation parameters as recommended by the American Association of Medical Physicists 120 (AAPM) Task Group 268 on the reporting of Monte Carlo radiation transport studies.Table 1 : Summary of the Monte Carlo simulation parameters used in this work.

[0066] Fig. 4(a) presents the backscattered fraction for monoenergetic electron beams directed towards a water phantom, as calculated by the backscatter clrp EGSnrc user-code. As expected, low energy electrons yield a higher fraction of backscattered electrons, with a backscattered fraction around 5.5% for energies around 50 keV. For the nominal energies considered in the testing (6 and 9 MeV), backscattered fractions on water are below 1%. Furthermore, the fraction of backscattered electrons for various materials at energies of 750 keV and 9 MeV is presented in Fig. 4(b). Results indicate that water and Acetal (polyoxymethylene), the material from which the Mobetron UHDR collimators are made, generate nearly identical fractions of backscattered electrons, around 1% at 9 MeV and less than 4% at 750 keV. For 9 MeV electrons, only lead induced a backscatter fraction above 10%.

[0067] Fig. 5 shows the total signal captured by the BCT as a function of SSD for 6 and 9 MeV UHDR beams directed towards a Solid Water® phantom. As observed, the BCT signal is reduced at shorter SSD, despite constant beam output. The effect is slightly higher for the 6 MeV beam, with a variation of 26.5% between 25 and 70 cm SSD, compared to 12.6% at 9 MeV.

[0068] The dependency of the BCT response as a function of the field size is presented in Fig. 6. The BCT signal is shown to increase with the collimator diameter, with variations as high 220 as 15% between 2.5 cm and 6 cm field sizes for both energies. This behavior, opposite to the effect of jaw setting and collimation size for monitor ion chamber in conventional linac (i.e. lower signal for smaller field size), was also observed in previous work.

[0069] The effect of phantom material on BCT signal was explored by comparing the responses from Solid Water® and water phantoms using a 9 MeV beam across fourvaried SSDs, ranging from 30 to 50 cm, as depicted in Fig. 7. A substantial difference in the BCT readings can be between the two phantoms, with discrepancies as high as 14.1% for 30 cm SSD. More specifically, the readings observed when using the water phantom are shown to be relatively unaffected by changes in SSD, maintaining consistent pulse shape and amplitude for all distances. In contrast, the Solid Water® phantom-induced BCT signal with noticeable drops within each pulse, an effect shown to be reduced as the phantom was placed further and further from the BCT. Water and Solid Water® are also shown to influence the BCT signal in different ways on a pulse-by-pulse basis, as reported in Fig. 8. First, it can be observed that all pulses yield a lower signal in Solid Water® compared to water, but the difference between the two phantoms is shown to increase after the first pulse. Indeed, the signal quickly reduces after the first pulse when delivered to Solid Water®, while the response is substantially more stable when delivered to water.

[0070] Fig. 9 provides more insights on how phantom properties might affect BCT signal. First, the plot in (a) shows that connecting a Solid Water® phantom to the ground noticeably affected the pulse shape and the total signal measured, despite identical incident beams. Indeed, the grounded Solid Water® results in a pulse shape that is flatter, akin to that seen in water, though not exactly to the same extent. This effect is also observed when a Roos chamber is either connected to or disconnected from the electrometer, with the connected chamber inducing a flatter pulse shape. These observations indicate that grounding a Solid Water® phantom, whether directly or via an ion chamber, substantially impacts the BCT reading.

[0071] The use of a Faraday shield 20 was tested as a mitigation strategy for BCT sensitivity against setup and irradiation parameters. This was achieved by placing the shield 20 directly at the exit of the test applicator, which was grounded to a Faraday cage on the head of the medical linear accelerator 10 using a conductive wire. For the clinical applicator, the design was slightly adjusted to hold an aluminium foil at a distance of 1.3 cm below the BCT 17 and above the collimator holder. All other characteristics of the applicator (material, length, diameter) were kept constant. Measurements for SSD, field size, and material dependence were repeated to assess the impact of Faraday shielding on BCT signal variations. T o achieve this, irradiations were duplicated with the shield 20 installed on the applicator, either grounded or ungrounded. This was done to specifically assess the effect of electrostatic shieldingon the BCT signal, without interference from the presence or absence of the shield 20 itself.

[0072] Fig. 10 presents the signal measured by the BCT for a 9 MeV UHDR beam at various SSD using Solid Water®, with the Faraday shield installed at the exit of the test applicator, either grounded or ungrounded. Results indicate that the Faraday shield removes essentially all SSD dependency when grounded, while the ungrounded shield generally reproduces what was reported in Figure 5, where no aluminium foil was used.

[0073] Similarly, Fig. 11 compares BCT signal for the same beam delivered to water and Solid Water® phantoms at a SSD of 30 cm when a grounded Faraday shield is installed on the applicator. Here again, the phantom material-specific responses and inter-pulse variability observed respectively in Figs. 8 and 9 are cancelled, resulting in virtually identical BCT response for water and Solid Water® when installing a grounded Faraday shield downstream of the BCT.

[0074] Finally, Fig. 12 illustrates the impact of field size on the BCT signal when a grounded Faraday shield is incorporated into our clinical applicator. With the shield in place and grounded, the BCT signal variation across the three field sizes remains below 0.5% for a given energy. In contrast, when the aluminum foil is ungrounded, the field size influences the BCT signal in a manner consistent with the observations made in Fig. 6.

[0075] In terms of the effect of irradiation conditions on BCT signal, with the objective of enabling reliable electron UHDR real-time beam monitoring, while several groups have validated the linearity of BCT signal as a function of pulse length and pulse rate frequency for UHDR electron beam using a constant setup, significant sensitivity of BCT signal (up to 12%) is exhibit for changes in SSD, field size, and phantom material despite constant beam parameters.

[0076] Monte Carlo simulations performed indicate that the backscatter fraction from a mono-energetic electron beam in water is less than 5% for most relevant energies and less than 1% for energies above 4 MeV. This suggests that even if all backscattered electrons were detected by the BCT 17 at short SSD and none were detected at large SSD, the difference in BCT signal between the two conditions should not be more than 2-3% for 6 MeV and 9 MeV beams. While the Mobetron medical linear accelerator 10 is known to have a low-energy component within its spectrumdue to the absence of beam steering magnet 18 (Figs. 3A and 3B), no realistic spectrum could induce relatively high variations (more than 20% difference between 25 cm and 70 cm SSD for 6 MeV). Similarly, as shown in Figure 4, the plastic used to make the Mobetron UHDR collimators, Acetal, induces less than 1% backscatter at 9 MeV, indicating that field size should only have a minor impact on BCT signal through backscattered electrons.

[0077] With the objective of identifying alternative causes for BCT signal variation as a function of SSD and field size, we compared signal obtained in Solid Water® to that measured in water with otherwise identical beam parameters. Results obtained in these conditions, presented in Fig. 7, show a dramatic impact of the phantom type on the BCT response, even though water and Solid Water® are considered equivalent in terms of MeV electron interaction properties, including backscatter. These results therefore suggest that a second phenomenon, unrelated to electron backscatter, substantially affected the BCT signal obtained using Solid Water®.

[0078] Since one of the main differences between water and Solid Water® is electrical conductivity, the effect could be related with the way charge dissipation occurs within the medium as the beam is delivered. Previous studies have demonstrated that some plastics (e.g. PM MA) could experience long-term charging following megavoltage electron irradiation at conventional dose rates and modern Solid Water® phantoms have been optimized to circumvent this limitation. However, at very high beam current, Solid Water® may still experience a transient charge loading phenomenon, where the electrons do not diffuse fast enough to reestablish charge equilibrium within the phantom. In that scenario, the build-up of negative charge in the irradiated region of the Solid Water® phantom would induce a growing electric field, a phenomenon already observed and documented for strongly insulating plastics under megavoltage electron irradiation.

[0079] To test if the difference between BCT signal captured for water and Solid Water® phantoms was effectively due to the different levels of charge dissipation in each medium, a 5 cm Solid Water® slab was irradiated, the Solid Water® slab placed above a thin aluminium foil grounded to earth through a conductive wire. Results, shown in Fig. 9, demonstrate that this indeed modified the BCT signal, even though all other irradiation parameters were kept constant. The shape of the pulse captured by the BCT for the grounded Solid Water® was closer to the one measured in water,but not perfectly alike, suggesting partial but incomplete resolution of the charge buildup effect through the grounded foil.

[0080] From this point, to evaluate if charge dissipation in medium did affect BCT signal through an electric field generated by transient charge build up during irradiation, Faraday shields 20 were added to prevent such an electric field from reaching the central region of the toroid. This was done on two applicators by placing a thin aluminium foil ( / .e., one possible type of shield 20) just below the BCT on which a grounded electric wire could be connected. Then, measurements were performed at various SSD, field sizes, and phantom materials with the foil grounded and ungrounded, to isolate the impact of the Faraday shielding from the simple addition of an aluminium foil downstream of the BCT. Results, reported in Figs. 10, 11 and 12, demonstrate that Faraday shielding was effective in removing virtually all SSD and field size dependence for the BCT measurements. Similarly, the shielded BCT yielded the same signal for Solid Water® and water for otherwise constant irradiation parameters, as one should expect.

[0081] Considering this, it can be asserted that the strong dependency of BCT signal as a function of SSD, field size, and phantom material is probably caused by an electric field originating in dielectric materials exposed to the UHDR electron beam. Indeed, for Solid Water® irradiations, the intensity of the electric field through an unshielded BCT would decrease as the SSD increases, which reflects the effect observed in previous work and reproduced in this study. Similarly, transient charge loading of the collimator material would be more important for smaller field sizes (larger surface of Acetal being irradiated), which is again in line with what has been reported in this work. Although both effects could also be explained by electron backscatter in terms of relation with SSD and field size, the Monte Carlo simulations indicate that the amplitude of the observed effects on BCT signal is too large to be explained by backscatter alone. Instead, by removing virtually all SSD, field size, and phantom material dependency through grounded Faraday shielding, our investigation provides strong evidence to support the hypothesis of an electric field originating in the irradiated material. To the best of our knowledge, this is the first time this effect has been identified and reported in the context of FLASH radiotherapy.

[0082] Observations provided above have important implications for the use of BCT to monitor UHDR electron beams. First, robust BCT calibration is probably not achievable if the effect of transient charge loading in media is not accounted for.Indeed, the BCT response presented in Fig. 7 shows that a dramatically different signal can be obtained for the same irradiation delivered to water and Solid Water®. Therefore, a BCT calibrated against a reference detector in Solid Water® is likely to report an inaccurate beam output if the beam is then used to irradiate a subject of different electrical conductivity (e.g. irradiation of cell cultures, organoids, a small animal or a human subject) if this effect is overlooked, as the electric field generated during calibration and beam monitoring might be drastically different. Similarly, using a BCT to correct beam output fluctuations for the cross calibration of an ion chamber and a passive detector (e.g. radiochromic film, alanine, etc.) in Solid Water® should be discouraged if a Faraday shield is not used, as the graphs herein indicate that the presence of a connected ion chamber can reduce the BCT signal perturbation caused by Solid Water®. Finally, for BCTs with a wide enough dynamic range to cover conventional and UHDR irradiations, outcome comparisons between the two regimes based on the BCT signal would likely be biased if the effect of electric fields is not taken into account.

[0083] Therefore, the BCT signals can be impacted by backscattered electrons and / or by electric fields generated by the transient charging of plastic materials under UHDR electron irradiation. The introduction of Faraday shielding 20 is a promising solution to mitigate the discrepancies in BCT signal across varying conditions and phantom materials.

Claims

CLAIMS1 . A medical linear accelerator configured to operate in radiotherapy, comprising: an electron gun for emitting an electron beam; an accelerator waveguide for accelerating the electron beam; a linear accelerator head defining an inner beam passage for directing the electron beam; a beam current transformer in the linear accelerator head for monitoring the electron beam; and at least one electrostatic shield in the inner beam passage, the electrostatic shield being positioned downstream of the beam current transformer for shielding the beam current transformer.

2. The medical linear accelerator according to claim 1 , wherein the linear accelerator head has a plurality of collimators.

3. The medical linear accelerator according to claim 2, wherein the beam current transformer is in one of the collimators.

4. The medical linear accelerator according to any one of claims 1 to 3, wherein the linear accelerator head includes an applicator.

5. The medical linear accelerator according to claim 4, wherein the beam current transformer surrounds or is placed within the applicator.

6. The medical linear accelerator according to claim 5, wherein the applicator has a downstream portion and an upstream portion, the downstream portion and the upstream portion interconnected to one another.

7. The medical linear accelerator according to claim 6, wherein the downstream portion and the upstream portion define a space to receive the beam current transformer therein.

8. The medical linear accelerator according to any one of claims 5 to 7, wherein the at least one electrostatic shield is in the applicator.

9. The medical linear accelerator according to any one of claims 5 to 8, wherein the at least one electrostatic shield is laid flat against at least one axial face of the beam current transformer.

10. The medical linear accelerator according to any one of claims 1 to 9, wherein the at least one electrostatic shield is a sheet.

11. The medical linear accelerator according to claim 10, wherein the at least one electrostatic shield is a sheet of conductive polymer.

12. The medical linear accelerator according to claim 11 , wherein the conductive polymer is one of polyimide, polyester and polyethylene.

13. The medical linear accelerator according to claim 10, wherein the at least one electrostatic shield is a metallic sheet.

14. The medical linear accelerator according to claim 10, wherein the at least one electrostatic shield is a graphene sheet.

15. The medical linear accelerator according to any one of claims 1 to 14, wherein the electrostatic shield lies in a plane across the inner beam passage.

16. The medical linear accelerator according to claim 15, wherein a central axis of the inner beam passage is normal to the plane.

17. The medical linear accelerator according to any one of claims 1 to 16, wherein the electrostatic shield is grounded to earth.

18. The medical linear accelerator according to any one of claims 1 to 17, comprising two of the electrostatic shield, one of the electrostatic shield being downstream of the beam current transformer, and another one of the electrostatic shield being upstream of the beam current transformer.

19. The medical linear accelerator according to any one of claims 1 to 18, further including a bending magnet configured to direct the electron beam in the inner beam passage.

20. A method for operating a medical linear accelerator in radiotherapy comprising: within the medical linear accelerator, emitting an electron beam, accelerating the electron beam, directing the electron beam via an inner beam passage,monitoring the electron beam in the inner beam passage by a beam current transformer, outletting the electron beam from the medical linear accelerator, and shielding the beam current transformer at least from backscatter.

21. The method according to claim 20, wherein directing the electron beam includes collimating the electron beam.

22. The method according to claim 21 , wherein collimating the electron beam includes collimating the electron beam in at least two collimators.

23. The method according to any one of claims 20 to 22, wherein directing the electron beam includes bending the electron beam to the inner beam passage.

24. The method according to any one of claims 20 to 23, including adjusting the emitting of the electron beam as a function of a value of the monitoring of the electron beam by the beam current transformer.

25. The method according to any one of claims 20 to 24, further including shielding the beam current transformer from energy accumulated in the medical linear accelerator upstream of the beam current transformer.

Images