Ionizing-Radiation Beam Monitoring System

- Integrated Sensors, LLC

A transmissive ionizing-radiation beam monitoring system includes an enclosure structure including an entrance window and an exit window to an incident ionizing-radiation beam, where the entrance window and the exit window are highly transmissive. The system further includes a thin scintillator within the enclosure structure that is directly in an incident ionizing-radiation beam path and transmissive to the incident radiation beam and an ultraviolet (“UV”) illumination source within the enclosure structure facing the scintillator for internal system calibration. Embodiments further include a UV photosensor within the enclosure structure positioned to monitor and calibrate the UV illumination source and a machine vision camera within the enclosure structure that includes a lens which views the scintillator through a close proximity mirror including a folded optical axis system located to a side of the scintillator.

Skip to: Description  ·  Claims  · Patent History  ·  Patent History
Description
CROSS-REFERENCE TO RELATED APPLICATIONS

This application is a continuation of U.S. patent application Ser. No. 16/529,200, filed on Aug. 1, 2019, which claims priority to U.S. Provisional Pat. Appln. Ser. No. 62/714,937, filed on Aug. 6, 2018, to U.S. Provisional Pat. Appln. Ser. No. 62/815,006, filed on Mar. 7, 2019, and to U.S. Provisional Pat. Appln. Ser. No. 62/859,952, filed on Jun. 11, 2019. The disclosure of each of these applications is hereby incorporated by reference.

GOVERNMENT LICENSE RIGHTS

The invention was made in part with government support under one SBIR (Small Business Innovation Research) Grant (Number: 5R44CA183437) awarded to Integrated Sensors, LLC by the National Institutes of Health (National Cancer Institute), and two SBIR Assistance Agreements (Award Nos. DE-SC0013292 and DE-SC0019597) awarded to Integrated Sensors, LLC by the U.S. Department of Energy (Office of Science). The government has certain rights in the invention.

FIELD

One embodiment is directed generally to radiation beam monitoring, and in particular to monitoring ionizing beams of particle or photon radiation while having minimal impact on the quality of the radiation beam itself.

BACKGROUND INFORMATION

The most common type of radiation therapy for the treatment of cancer is external beam radiation therapy (“EBRT”). For this treatment, an accelerator is used to generate and precisely deliver relatively high-energy particle or photon beams from outside the body into the tumor. There are a variety of EBRT technologies, with the type of radiation used falling into two general categories: (1) ionizing particles such as protons, ions, electrons, etc., and (2) ionizing photons such as relatively low-MeV gamma rays or X-rays. Ionizing photons are the more common type of radiation employed for EBRT. For particle beam radiation therapy, in addition to protons, carbon ions and electrons, other types of particle beams used or being investigated include helium, oxygen, neon and argon ions, as well as low-energy neutrons (e.g., slow to thermal neutrons). Low-energy neutrons are used, for example, in boron neutron capture therapy (“BNCT”) and gadolinium neutron capture therapy (“Gd-NCT”).

For both particle and photon EBRT, there are a variety of delivery methods, including intensity modulated radiation therapy (“IMRT”), intensity modulated proton therapy (“IMPT”), three-dimensional conformal radiation therapy (“3D-CRT”), image guided radiation therapy (“IGRT”), volumetric modulated arc therapy (“VMAT”), pencil-beam spot scanning, pencil-beam raster scanning, helical-tomotherapy, stereotactic radiosurgery (“SRS”), stereotactic body radiation therapy (“SBRT”), fractionated stereotactic radiotherapy (“FSRT”), spatially fractionated grid radiation therapy (“SFGRT”), ultrahigh dose-rate flash therapy (“FLASH”), intraoperative radiation therapy (“IORT”), boron neutron capture therapy (“BNCT”), gadolinium neutron capture therapy (“Gd-NCT”), etc.

SUMMARY

One embodiment is a transmissive ionizing-radiation beam monitoring system that includes an enclosure structure including an entrance window and an exit window to an incident ionizing-radiation beam, where the entrance window and the exit window are highly transmissive. The system further includes a thin scintillator within the enclosure structure that is directly in an incident ionizing-radiation beam path and transmissive to the incident radiation beam and an ultraviolet (“UV”) illumination source within the enclosure structure facing the scintillator for internal system calibration. Embodiments further include a UV photosensor within the enclosure structure positioned to monitor and calibrate the UV illumination source and a machine vision camera within the enclosure structure that includes a lens which views the scintillator through a close proximity mirror including a folded optical axis system located to a side of the scintillator.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a radiation damage recovery plot as a function of time (in hours) for 191 μm thick BoPEN film in accordance to embodiments.

FIG. 2 is a plot of the average pixel signal decrease as a function of time for a 191 μm thick BoPEN film in accordance to embodiments.

FIG. 3 is a plot of the fluorescence light loss and recovery in air as a function of time for different thicknesses of BoPEN film in accordance to embodiments.

FIG. 4 illustrates an example of the Spread-Out Bragg Peak (“SOBP”) for an X-ray photon vs. proton beam in accordance to embodiments.

FIG. 5 is a plot showing the exponential fluorescence decrease as recorded by the average camera pixel signal in accordance to embodiments.

FIGS. 6A-B illustrate two images of a 10 nA, 3.0 MeV proton beam inside a vacuum chamber in accordance with embodiment.

FIG. 7 is a projection of the camera field-of-view for the digital image in FIG. 6A, taken at a working distance of 326 mm in accordance with embodiments.

FIGS. 8A-C illustrate a system that includes a two camera, single scintillator beam monitor in a light-tight enclosure employing a rolled scintillator spool configuration in accordance to embodiments.

FIGS. 9A-D illustrate a system that includes a two camera, single scintillator roll film beam monitor with linear translation of the scintillator spool system in a 6-way-cross vacuum chamber in accordance with embodiments.

FIGS. 10A-C illustrate a system that includes a roll film scintillator beam monitor in a smaller 6-way-cross vacuum chamber without linear translation capability in accordance with embodiments.

FIGS. 11A-D illustrate a system that includes a single scintillator-frame beam monitor in 6-way-cross vacuum chamber in accordance with embodiments.

FIGS. 12A-C illustrate a system that includes a double scintillator-frame beam monitor in a 6-way-cross vacuum chamber in accordance with embodiments.

FIGS. 13A-C illustrate a system that includes a double scintillator-frame beam monitor in a 6-way-cross load-lock vacuum chamber in accordance with embodiments.

FIGS. 14A-D illustrate a system that includes a two camera, two mirror, full-size single scintillator/window module beam monitor in a slim light-tight enclosure in accordance with embodiments.

FIGS. 15A-C illustrate a system that includes a one camera, one mirror, half-size rectangular single scintillator beam monitor in accordance with embodiments.

FIGS. 16A-C illustrate a system that includes a three camera version of the embodiments shown in FIGS. 15A-C in accordance with embodiments.

FIGS. 17A-B illustrate a system that includes a four camera version of the embodiments shown in FIGS. 14A-D for the full-size single scintillator-frame beam monitor in accordance with embodiments.

FIGS. 18A-B illustrate a system that includes a four camera, full-size double window/scintillator module beam monitor in a light-tight slim enclosure in accordance with embodiments.

FIGS. 19A-B illustrate a system that includes an eight camera, full-size double window/scintillator module beam monitor in a light-tight slim enclosure in accordance with embodiments.

FIGS. 20A-C illustrate a system that includes a four camera, single scintillator beam monitor employing a rolled scintillator spool configuration in accordance with embodiments.

FIGS. 21A-B illustrate a two camera, full-size single scintillator-frame beam monitor without mirrors in a light-tight box enclosure in accordance with embodiments.

FIGS. 22A-C illustrate a system that is a four camera version of FIGS. 21A-B in accordance with embodiments.

FIGS. 23A-B illustrate 10 μs exposure camera images through a vacuum chamber window in accordance to embodiments.

FIG. 24 illustrates a 1 ms exposure of a captured image of a ˜2 mm diameter proton beam irradiating an ultra-thin 12.2 μm BoPEN film while moving back and forth in a rastered zig zag pattern at 40 mm/ms in accordance to embodiments.

FIG. 25 illustrates a four plate light baffle for air circulation in accordance to embodiments.

FIG. 26 is a photograph of a 2×2 inch rectilinear image taken at a 45° tilt angle in accordance to embodiments.

FIGS. 27A-B illustrate the open central structure of a reduced/shortened 4″ O.D. tube, 6-way-cross with 6″ diameter CF-flanges modified such that the total beam entrance-to-exit length is 5.9″ in accordance to vacuum chamber embodiments.

DETAILED DESCRIPTION

Embodiments are directed generally to ultra-fast transmissive (“UFT”) two-dimensional (“2D”), high resolution, ionizing particle and photon beam monitors primarily for applications based on, or related to, external beam radiation therapy (“EBRT”), including the monitoring in “real-time” of beam position and movement, intensity profile including tail, beam fluence/external dosimetry, angular divergence and patient treatment quality assurance.

In embodiments of the present disclosure, the term “ultra-fast” refers to “real-time” on-line monitoring and data analysis of streaming images of an ionizing-radiation beam within approximately 10 ms or less per image, corresponding to a data analysis rate of approximately 100 frames per second (“fps”) or faster. For some embodiments, the streaming images can be coming in at rates of 1,000 to 10,000 fps (i.e., 1 ms to 0.1 ms) with the data analysis occurring concurrently. Further, the terms “transmissive” and “highly transmissive” are adjectives used to describe the relatively small amount of energy that a particle or photon loses in transit through a given material or system, which will be different for an entrance or exit window as compared to the scintillator material itself as compared to the integrated beam monitor system comprising the entrance window, exit window, scintillator, and the column of air between the entrance and exit windows. For any given system the relative amount of energy loss will vary greatly at different incident particle or photon energies which can vary over many orders-of-magnitude, and for different types of particles from neutrons to protons to carbon-ions, etc. For an EBRT application such as proton therapy using a proton beam having an incident energy of 210 MeV, the term “highly transmissive” would mean losing no more than about ≤0.1%) of its incident energy in transit through the UFT beam monitor system (i.e., losing ≤0.2 MeV), but for the exact same system at 80 MeV “highly transmissive” would mean losing ≤0.5%) (i.e., losing ≤0.4 MeV). For this example at the same energies, the term “transmissive” would mean at 210 MeV losing no more than about ≤0.2%) of its incident energy in transit through the UFT beam monitor system (i.e., losing ≤0.4 MeV), but for the same system at 80 MeV the term “transmissive” would mean losing 1% (i.e., losing ≤0.8 MeV).

The beam monitors in accordance to embodiments incorporate thin and ultra-thin scintillator materials (e.g., scintillator sheet or film material) and are capable of internal, frequent, self-calibration to compensate for a variety of factors including system non-uniformity including camera sensor/pixel response, optical system distortions, slow degradation of the scintillator material due to radiation damage, signal drift due to temperature rise within the monitor enclosure, etc. In embodiments, the term “ultra-thin” refers to both window (i.e., entrance and/or exit window) and scintillator materials having a thickness of ≤0.05 mm, and the term “thin” refers to scintillator materials having a thickness of ≤0.5 mm and thus also includes ultra-thin scintillators.

The integrated detector/monitor in accordance to embodiments has an intrinsic 2D position resolution in the range of ˜0.03 mm to 0.2 mm, depending on the application specification requirements, and is highly transparent to the incident ionizing particle or photon beam, thereby resulting in minimal beam scatter, low to extremely low energy straggling, and minimal generation of secondary radiation. Embodiments, in addition to EBRT, can be used for the monitoring of low-luminosity exotic particle beams and/or high-luminosity particle beams generated by research accelerators for scientific experiments, industrial particle and photon beam monitoring for materials processing (e.g., high energy ion implantation, food and medical sterilization, cutting and welding, etc.), materials analysis, non-destructive analysis, radioisotope production, etc. Beam monitors in accordance to embodiments generally do not require a controlled atmosphere or vacuum environment for proper operation, although some embodiments have been designed for operation in vacuum or controlled gaseous environments.

Embodiments for EBRT applications generally result in positioning the beam monitor downstream from the accelerator exit nozzle in an ambient air atmosphere. However, other embodiments are configured to operate within the vacuum environment of the beamline pipe to optimize and/or monitor the beam shape, intensity, position and beam focus prior to reaching the beam exit/nozzle or target region. Embodiments for EBRT applications downstream from the nozzle incorporate a unique folded optical configuration to achieve a thin profile to minimize encroaching upon the confined and narrow space between the beam nozzle exit and the patient.

Due to the ultra-fast response capability of the embodiments of beam monitors, they can provide sub-millisecond and even microsecond beam analysis and feedback to the delivery system, thereby allowing corrective actions to be taken if necessary. For EBRT, this capability can potentially improve the treatment delivery efficacy and protect the patient, especially for recent “FLASH” therapy applications. For nuclear and high energy physics, this capability can provide particle time-of-flight (TOF) information in the range of 50 to 100 μs, or greater.

It is known to use a scintillator, including a plastic scintillator to detect ionizing radiation, coupled with an electronic photodetection device to quantitatively measure the emitted photons from the scintillator. It is also known to use a digital camera to record the light emitted from an irradiated scintillator in applications ranging from monitoring the beam shape and position of an electron beam, to using X-rays irradiating a scintillator to evaluate the quality of mechanical welds, to optimizing the beam delivery system used in proton beam therapy.

In contrast to known uses, embodiments implement multi-camera folded optical configurations, such as 2, 3, 4, 6, 8, 10, 12 cameras, for advanced beam monitoring systems that provide critical performance and space-saving advantages such as extremely high spatial resolution while minimizing encroachment on the limited space existing between the EBRT exit nozzle and the patient's body. Embodiments also include configurations of relatively compact machine vision cameras with imaging sensors that can stream images live to a computer system that includes a frame grabber for real-time data processing and analysis, the use of machine vision cameras that can be programmed for application specific parameter optimization such as selection of exposure time, gray scale level (i.e., bit depth), acquisition control and frame rate, gain control, black level control, gamma correction for pixel intensity, pixel binning, pixel sharpening, windowing down the area or region of interest to achieve higher frame rates for faster beam analysis. Embodiments further include the use of both single and double scintillator configurations that can be integrated as part of an easy to replace foil-window/scintillator module package, and rolled scintillator-film motorized spool assemblies for automated scintillator film advancement/replacement that uses novel polymer thin film scintillator materials such as biaxially-oriented polyethylene naphthalate (“BoPEN”), biaxially-oriented polyethylene terephthalate (“BoPET”), polyethersulfone (“PES”), etc. that are intrinsic scintillators without the addition of fluor dopants. Embodiments include novel designs for quick replacement of radiation damaged scintillator film or sheet with new scintillator film or sheet without significant service downtime and recalibration time associated with the scintillator replacement process, configurations for real-time beam monitoring systems operating in a vacuum environment, configurations for beam monitoring systems operating in either a naturally circulating or controlled flow-through ambient air or special gaseous environment such as an enriched oxygen gaseous atmosphere to possibly minimize radiation damage by enhancing oxygen assisted radiation damage recovery, configurations incorporating actively cooled camera sensors for enhanced performance and reduced radiation damage of the camera sensor element, configurations incorporating the addition of internal UV sources such as UV-LEDs and internal UV detectors such as UV-photodiodes and appropriate filters such as bandpass filters to achieve internal self-calibration of system non-uniformity and near continuous self-correction for progressive scintillator radiation damage; real-time software correction of optical system distortions, perspective distortions (e.g., keystoning), aberrations and non-uniformities including camera image sensor pixel defects and non-uniformity. Embodiments include configurations utilizing 3-way tees or wyes, 4-way-cross, 5-way-cross and 6-way-cross vacuum chamber configurations for beamline vacuum operation that allow the use of either two cameras, or two photomultiplier tubes (“PMT”s), or one camera and one PMT, or PMT replacements such as solid state photomultipliers (“SSPM”) including silicon photomultipliers (“SiPM”), avalanche photodiodes (“APD”), single-photon avalanche diodes (“SPAD”), etc. Embodiments include high dynamic range (“HDR”) computational imaging and with the thinnest scintillator films have extremely low beam energy straggling with minimal generation of secondary ionizing particles and photons.

Embodiments achieve advantages in part by using a scintillator film material, available in continuous rolls (e.g., >1000 ft length) of about 70 cm width and greater, and thicknesses from about 1 μm to 250 μm in conjunction with other components to achieve unexpected results with regard to radiation damage resistance, photon emission, and as a thin and/or ultra-thin film scintillator. Embodiments include designs to take advantage of the new thin and ultra-thin scintillator material which is highly resistant to radiation damage, while being able to minimize and possibly eliminate most problems having to do with scintillator non-uniformity and time consuming scintillator material replacement and system calibration.

Embodiments include an innovative folded-optics design to minimize the product profile/thickness to within about 6-14 cm, depending upon scintillator and camera size and camera angle. Embodiments include an innovative automated, internal, rapid calibration system using UV-LEDs, UV-photodiodes, and UV and VIS bandpass filters, with an estimated time for system calibration of about one minute or less. Embodiments include machine vision cameras that would typically stream images at frame rates from about 100 fps to 40,000 fps.

Embodiments discussed below include an in-line beam monitor design (e.g. FIGS. 11-13) with fast, high gain photomultipliers (e.g., approaching 1×107), coupled with an efficient photon collection system and suitable scintillator and radiation source (e.g., highly ionized particles with an atomic number of ˜10 or greater, such as Ne+10) capable of generating at least ˜200 photoelectrons and achieving on the order of about 100 μs timing resolution, and possibly better than 50 μs timing resolution, which is critically important for time-of-flight (“TOF”) experiments.

Embodiments further enhance timing resolution for TOF measurements by increasing photon collection, such as through the use of two PMTs or SSPMs in the opposite arms of a 6-way-cross instead on one PMT (or SSPM) and one camera, or improving the collection of photons from the front side of a scintillator by depositing a reflective coating on the scintillator back side, or roughening the front collection surface of a scintillator to prevent total internal reflection.

Embodiments include multi-camera configurations (e.g. 2, 3, 4, 6, 8, 10, 12, etc., cameras) with FPGA frame grabbers and software that can compile, integrate and analyze streaming images in real-time of the moving beam, while correcting for optical image perspective/keystone distortions, lens distortions, vignetting, scintillator non-uniformity, camera sensor pixel non-uniformity, defective and radiation damaged pixels, etc.

Embodiments include manual or motor controlled push-pull linear positioners and/or rotary drives to advance fresh scintillator film as needed into the incident beam active area. Embodiments include a load-lock vacuum chamber design to change scintillator films without having to break the beamline vacuum. Embodiments include an ultra-thin, light-blocking beam entrance and exit foil and/or polymer window, bonded to a thin frame, that can also be bonded to the scintillator film or sheet material to make a simple window/scintillator replaceable module package that can be dropped into a pocket in the beam monitor front and/or back cover plate and calibrated within a minute or so without having to open up the system enclosure.

Embodiments have a design based on two different in-line scintillators, one sensitive to essentially all particles and high energy photons/gammas except neutrons, and the other doped with a high neutron cross-section isotope such as B10, Li6 or Gd in order to make it neutron sensitive. By digitally subtracting the image/signal of the first scintillator from that of the second scintillator, the resulting second scintillator image/signal will be primarily that of the neutron beam and can achieve the high performance at low cost desired in a high gamma discrimination neutron detection system.

Most known EBRT particle accelerators are designed for pencil-beam spot scanning, but a few systems are designed for pencil-beam raster scanning. The beam monitor embodiments disclosed below are compatible with both types of pencil-beam scanning systems, with most configured to operate downstream from the exit nozzle, but some embodiments have been designed to operate upstream of the nozzle in the vacuum environment of the beamline delivery system either in the patient treatment room or prior to the treatment room and switch house and close to the accelerator. The purpose of such systems operating in the beamline vacuum is usually diagnostic to facilitate beam tuning including measurement and optimization of the 2D beam profile in the delivery system, whether for EBRT, or for nuclear and high energy physics. In all cases, the scintillator material should be an extremely thin film so as to be almost transparent with very little low energy straggling so as not to degrade the beam in the process of measuring it. For such applications, the scintillator film in some embodiments should be less than 100 μm thick and possibly as thin as 1 μm.

In embodiments the scintillator film BoPEN is employed in thickness down to 1 μm, and in some embodiments this film is physically attached to a rigid frame as shown in some of the 6-way-cross embodiments disclosed below.

Experimental results showed a 33.0% initial decrease in scintillator fluorescence from a 191 μm thick BoPEN film measured almost immediately after being irradiated for 5 minutes by a low energy proton beam that resulted in a film dose of 59 kGy. Specifically, FIG. 1 is a radiation damage recovery plot as a function of time (in hours) for 191 μm thick BoPEN film exposed at a proton dose rate of 0.20 kGy/s, for 5 minutes, corresponding to a total dose of 59 kGy in accordance to embodiments. In the left plot the relative light loss was measured almost immediately in air after exposure, while in the right plot the sample was kept for ˜21 hours in vacuum before being removed and then measured in air. However, after a 19-hour period in an ambient air atmosphere, the fluorescence emission in the left plot had partially recovered and the decrease was measured to be about 19.7%, representing about a 40% radiation damage reversal/recovery in less than one day. As shown in FIG. 1, most of the recovery occurred within the first few hours, but radiation damage recovery in air can continue for many days and even weeks, albeit at a much-reduced rate.

The above BoPEN film (density of 1.36 g/cc) radiation damage experiments employed a 5.4 MeV proton beam that was continuously irradiated for 300 seconds at an incident beam current density of 2.4 nA/cm2. Upon passing through the BoPEN film, each 5.4 MeV proton loses about 2.14 MeV. Given the above beam current and integrated exposure time, the BoPEN scintillator film was subjected to 59 kGy of accumulated radiation dose (1 Gy=1 J/kg) absorption as calculated below:


N=(2.4×10−9 coul/sec-cm2)*(6.25×1018 protons/coul)*(300 sec)=4.5×1012 protons/cm2


J=2.14 MeV/proton=(2.14×106 eV/proton)*(1.6×10−19 J/eV)=3.4×10−13 J/proton


Mass=(1.36 g/cm3)*(0.0191 cm)=0.026 g/cm2=2.6×10−5 kg/cm2


Dose=(3.4×10−13 J/proton)*(4.5×1012 protons/cm2)/2.6×10−5 kg/cm2=1.53J/2.6×10−5 kg=59×103 Gy


Dose rate=59 kGy/300 sec=0.20 kGy/s

For a transmissive particle beam monitor based on viewing and measuring the beam via its effect on scintillator emission, the effect of radiation damage can be quantified by equating it to the reduction in scintillator yield as measured by relative light loss (i.e., fluorescence signal reduction). With respect to obvious visual radiation damage, none of the more than 30 BoPEN samples irradiated to date, at dosage levels up to ˜60 kGy, have shown any visual signs of scintillator discoloration or surface degradation which was an unexpected result. However, BoPEN films have discolored at 400 kGy (see discussion below and Table 1).

In estimating the acceleration factor in experimental uses for different applications, with proton beam therapy being of particular interest, an average conventional daily patient treatment regime delivers ˜2 Gy per session. So, the above test that delivered 59,000 Gy to the BoPEN scintillator film in 300 seconds is presumably equivalent to the dose incurred in treating ˜30,000 patients. In other words, 1 second of accelerated irradiation in the 5.4 MeV test beam, approximately simulates the radiation received by the scintillator in conventionally treating ˜100 patients (a lesser number of patients for FLASH therapy). Or viewed another way, if a typical proton beam treatment room can process about 30 patients per day, then 5 minutes of the above accelerated proton beam test is equivalent to ˜1000 days of conventional patient treatments in a one-room facility. This degree of radiation damage resistance, with no obvious visual sign of surface degradation or discoloration in an off-the-shelf commercial polyester film, under such an aggressive, high rate, accelerated testing regime is an unexpected result.

FIG. 2 is a plot of the average pixel signal decrease as a function of time for a 191 μm thick BoPEN film exposed at a proton dose rate 9.2 kGy/s, for 53 seconds, corresponding to a total dose of 490 kGy in accordance to embodiments. The time scale is shown in terms of the camera image numbers recorded at 2 fps of the BoPEN film while being irradiated in a vacuum chamber. It can be seen that even at this high dose rate there is relatively little difference between the linear and exponential fits.

From the BoPEN scintillator test experimental results to date, it appears that at the above dose rate of 0.2 kGy/s that radiation damage is linear with exposure, up to rates approaching 10 kGy/s (as shown in FIG. 2). By assuming linearity, corrections can be made for the increased radiation damage that occurs at the sample back surface by supposing that the average value between the front entrance and back exit surfaces provides a reasonable estimate of the fluorescence coming from the sample middle bulk layer. In making this correction, the bulk fluorescence value from the sample center now becomes 22.2% greater than from the back surface. Thus the measured back surface light loss values can be corrected to obtain a more accurate bulk fluorescence value by multiplying the measured back surface values in FIG. 1 by 77.8%. Making this correction, the 59 kGy dose exposure which caused a 33.0% decrease in initial fluorescence and the 19.7% decrease in fluorescence after 19 hours, respectively becomes a 26% decrease (i.e., 0.778*33%) in initial fluorescence and a 15.6% decrease after 19 hours.

Hence the measured rad-damage values previously stated and appearing in FIG. 1 are overstated by 22.2% for the 191 μm thick BoPEN films. However, for ultra-thin BoPEN films with a thickness of ≤25 μm, the correction factors would be <1% and negligible. Likewise, for therapeutic particle beams with incident proton energies in the range of 70-225 MeV, even for the thin 191 μm thickness BoPEN film, there is no significant difference between the dose received at the front and back layers, and thus the surface fluorescence signal as measured using a 280 nm UV-LED source (located behind the BoPEN film and 99% absorbed in the first 0.1 μm back surface layer) accurately represents the radiation damage at these energies to the bulk material.

FIG. 3 is a plot of the fluorescence light loss and recovery in air as a function of time for different thicknesses of BoPEN film, ranging from 3.0 μm to 191 μm, after exposure to a 5.4 MeV proton beam in accordance to embodiments.

As can be seen in FIGS. 1 and 3, and in general for all BoPEN films, most of the radiation damage recovery in air occurs within the first few hours. The differential radiation damage effect described above, for thick versus thin scintillator films, explains the apparent differences in relative light loss shown in FIG. 3. For example, in comparing the maximum 7.3% light loss for the 12 μm thick BoPEN film measured after 40 minutes to the 17.5% light loss (from the back surface) for the 191 μm thick BoPEN film plotted after 40 minutes (i.e., dashed vertical line at 0.67 hours), the two values agree within an uncertainty of ±10% (i.e., 7.3% vs. 7.9%). The calculation for 191 μm thick BoPEN film is as follows:


Initial adjustment for front surface light loss in 191 μm film=(1−0.444)117.5%)=9.7%


Additional adjustment for difference in beam current density=(1.10/1.35)*(9.7%)=7.9%

Further, in experiments there was excellent agreement from two different 191 μm thick BoPEN samples, measured more than three weeks apart, using significantly different beam currents. In particular, the initial 19.7% light loss in FIG. 3, when adjusted for the higher beam current density in FIG. 1 (2.4/1.35), yields an adjusted beam loss value of 35.0% compared to 33.0% in FIG. 1.

In estimating the beam energy lost in transit through the film, and the beam shape and intensity via its fluorescence profile, it is necessary to know the BoPEN film thickness and uniformity. A convenient non-destructive method for measuring film thickness and uniformity is via the front/back surface reflectance generated by spectral interference. This method can accurately measure film thicknesses over the full range from ˜1 μm to 250 μm, and to within about ±0.1 μm accuracy. For the films in FIG. 3, the measured thicknesses were: 3.0, 5.8, 12.2 and 191.0 μm, as measured by spectral reflectance in the near-IR over the wavelength range from ˜1,000 to 1,900 nm.

The above data indicates that thinner BoPEN scintillator films appear to be more radiation damage resistant than the thicker films (e.g., see the 300 second plots for the 3, 6, 12 and 191 μm thickness BoPEN films in FIG. 3 at both 0.67 and 2.45 hours). This result is completely unexpected and surprising, and counter intuitive to what was previously expected. Explanations for this unexpected result include that radiation damage recovery is significantly faster in the thinner films in air as suggested by the narrow dashed vertical line intersecting the thickness film plots at 2.45 hours and the projections further out at 4-5 hours in which the radiation damage appears to be essentially fully reversed for the thinnest films as compared to the 191 μm thick film. Another explanation includes that radiation damage depends on the probability of free-radical interaction, or a multi-particle free-radical mechanism, and so the thicker films having a greater free-radical density at the exit surface due to dE/dx, also has a higher probability of single or multiple free-radical proximity interactions. Other explanations include that the thinner films have a faster and higher probability of free-radical migration and diffusion to the film air surface, and because the FIG. 3 measurements were all in air, the thinner films have a higher rate of oxygen permeation and diffusion, as well as singlet oxygen escape. Being able to provide a verifiable mechanism to explain the higher radiation damage resistance of the thin films is not necessary, as the unexpected good news is that for BoPEN the thinner scintillator films appear to be more radiation damage resistant (i.e., rad-hard) than the thicker films.

In embodiments, in order to refine the above estimates for the scintillator dose exposure under more realistic clinical proton therapy conditions, an additional 20% scintillator dose can account for patient planning and calibration activity, and weekly machine maintenance. This adjustment means that the previously stated estimate of 30 patients per day, at 2 Gy per patient, corresponding to 60 Gy per day scintillator dosage, might prudently be increased by about 20% to 72 Gy per day. Therefore, the above calculated 59 kGy of accelerated exposure at a test facility, would be equivalent to 819 days of accumulated patient service assuming conventional irradiation treatment (i.e., not FLASH).

If a proton beam facility operates 5 days per week, then 819 days of service corresponds to 164 weeks which would be more than 3.1 years of continuous service. Assuming a linear radiation damage model (e.g., shown in FIG. 3), the previously revised measurement of a 15.6% loss in scintillator efficiency due to radiation damage (i.e., loss in fluorescence after 19 hours) would correspond to a 0.156% efficiency loss every 8.19 days. However, since radiation damage recovery continues well beyond 19 hours, then over the course of 8.19 patient treatment days the accumulated damage will certainly be less than 0.15%. More specifically, for a 5-day patient treatment week the accumulated proton rad-damage is likely ≤0.09% per week for the 191 μm thick BoPEN film assuming a 2 Gy dose to the scintillator per patient.

FIG. 4 illustrates an example of the Spread-Out Bragg Peak (“SOBP”) for an X-ray photon vs. proton beam in accordance to embodiments. FIG. 4 shows that the above estimate overstates the rad-damage to the scintillator, because 2 Gy to the patient does not equal to 2 Gy to the scintillator due to the SOBP. The SOBP means that if the tumor receives 2 Gy, then depending upon such factors as the tumor density, thickness and location, which determine the proton beam energy, the radiation dose delivered to the skin, or scintillator will typically fall in the range of about 50% to 75% of the dose to the patient's tumor, and would be about 1.0-1.5 Gy. Therefore, instead of the accumulated BoPEN rad-damage being ≤0.09% per week as estimated above, after correcting for the SOBP, the 191 μm thick BoPEN scintillator should suffer a rad-damage loss of only about 0.04% to 0.07% per week. This result is most surprising and leads to the unexpected conclusion that on a weekly and probably monthly basis the rad-damage to a 191 μm thick BoPEN scintillator is practically negligible, and even more so if a thinner BoPEN scintillator can be used.

FIG. 5 is a plot showing the exponential fluorescence decrease as recorded by the average camera pixel signal measured off of a 191 μm thick BoPEN film exposed at a proton dose rate of 460 kGy/s, for 33 seconds, corresponding to a total dose of 15,000 kGy in accordance to embodiments. The time scale is shown in terms of the camera image numbers recorded at 2 fps of the BoPEN film while being irradiated by a 3.0 MeV proton beam in a vacuum chamber. As shown in FIG. 5, and Table 1 below, at the highest measured dose rates, the radiation damage is not linear with exposure but is exponential. This means that estimates made at these high dose rates (e.g., ≥90 kGy/s) can be misleading in projecting greater scintillator radiation damage than actually incurs at lower dose rates integrated over longer periods of time. In other words, if the accelerated test data has a significant exponential component, then the actual scintillator radiation hardness would be better than stated above.

However, it appears that in the range of accelerated radiation dose rates chosen for modeling the performance of both therapeutic particle beams and for nuclear physics particle beam monitors (i.e., dose rates of 9 kGy/s shown in Table 1), projections based on a linear model should provide a good estimate of scintillator performance and any corrections for exponential behavior would be minor as shown in FIG. 2. For example, at a measured dose rate of 9.2 kGy/s, the linear relationship for BoPEN films still appears reasonable (see FIG. 2) corresponding to delivering 0.5 MGy in just 53 seconds (shown in Table 1), which is an unexpected result. Yet at dose rates of ≥90 kGy/s, deviations from linearity are major and calculations based on the assumption of linearity would be erroneous and should only be used for qualitative purposes. At these much higher dose rates, slow to moderate scintillator ablation begins instantaneously (see FIG. 5 and Table 1).

TABLE 1 Summary of MIBL Proton Beam Accelerated Test Results for 191 μm thick BoPEN Scintillator Dose Current Beam Beam Rate Density Current Energy Dose (kGy/s) (nA/cm2) (nA) (MeV) (kGy) Rad-Damage Observations Date 0.11* 1.35 5.4 5.4 33 No discoloration. Minimal rad-damage, largely reversible* Dec. 18, 2018 0.20 2.4 9.6 5.4 59 No discoloration. Fluorescence loss mostly reversible Dec. 18, 2018 3.3 40 10.0 5.4 390 Sample 16: Area darkening disappeared 2 months later Nov. 26, 2018 9.2 50 1 3.0 490 Sample 13: No ablation but 0.6%/sec fluorescence decrease Dec. 18, 2018 92 500 10 3.0 6,100 Sample 13: Immediate but slow surface ablation Dec. 18, 2018 460 2,500 50 3.0 15,000 Sample 13: Immediate fast surface ablation => deep hole Dec. 18, 2018 *Rate of 110 Gy/s with minimal rad-damage is well in excess of the 40 Gy/s rate used for FLASH proton therapy.

None of the BoPEN films receiving dosages up to 59 kGy (i.e., 300 seconds with 5.4 MeV proton beam at a current density of 2.4 nA/cm2) and at a dose rate of 0.20 kGy/sec showed any sign of surface degradation or discoloration despite significant decreases in fluorescence due to rad-damage as shown in FIGS. 1 and 3. Both figures show that measurable radiation damage recovery begins in air almost immediately after exposure and that this recovery can continue for days or even weeks afterwards. However in a vacuum environment, such recovery is either greatly reduced or delayed as shown in FIG. 1.

Rad-damage induced darkening (i.e. yellow-brown discoloration) has been observed in a 191 μm thick BoPEN film using the 5.4 MeV proton beam at a 10 nA current, with a fixed, non-rastered beam focused on a 0.25 cm2 area for 118 seconds. The resulting current density of 40 nA/cm2 yielded a dose rate of 3.3 kGy/s and produced an accumulated dose of 390 kGy. This dose rate was 16 times greater than received by the 59 kGy dose irradiated sample disclosed above. However, when the 390 kGy dose film was viewed two months later, it was discovered that the darkened/discoloration area had completely disappeared, so apparently at least some visually damaged BoPEN films can self-heal/recover in air to the extent that they no longer appear visually discolored.

In order to evaluate the dosage associated with irreversible physical damage (such as burning a hole into the film by proton ablation), a more stable fixed proton beam accelerator was used with a 191 μm thick BoPEN film at a reduced proton kinetic energy of 3.0 MeV and with a much tighter beam focus over an ablated hole area of 0.020 cm2 (i.e., diameter at hole surface was 1.6 mm as disclosed below), at beam currents of 1 nA for 53 seconds, 10 nA for 66 seconds, and 50 nA for 33 seconds. At each beam current a series of images were recorded at a shutter time/exposure of 1 ms, and at a frame rate of 2 fps for all three cases.

FIGS. 6A-B illustrate two images of the above 10 nA, 3.0 MeV proton beam, having approximately a 2.68 mm diameter, irradiating a 191 μm thick BoPEN film inside a vacuum chamber in accordance with embodiment. FIG. 6A is the digital image recorded with a 1 ms exposure and the pixel image resolution is 38.2 μm. FIG. 6B is a Gaussian fit to FIG. 6A with a measured average a=0.61 mm and 97% of the beam falling within a 2.2σ radius of 1.34 mm.

The number of images recorded for the above experiment corresponded to 89 images at 1 nA (see disclosure below), 133 images at 10 nA, and 67 images at 50 nA, with the fluorescence pattern and signal intensity recorded for each picture on a pixel-by-pixel basis as seen in FIGS. 6A and 6B for the 1st image taken at a beam current of 10 nA. A complete set of pictures at the three beam currents were taken without breaking vacuum or moving the camera, and by sequentially increasing the beam current after each set of images (i.e., from 1 nA, to 10 nA, to 50 nA) while the beam remained focused on the same scintillator spot area. Therefore, when the BoPEN film was finally removed after the last image at 50 nA, the partially ablated hole/crater represented the sum total from the three beam current doses piled on top of one another. Although the 1 nA beam caused no obvious physical film damage, it did suffer a 0.6% decrease in fluorescence per second of irradiation (i.e., slope was 0.003, see FIG. 2).

As previously disclosed, the fluorescence decrease followed close to a linear fit as seen by the solid line in FIG. 2 (there were two dead periods when images were not taken but beam exposure continued); however, the dotted line in FIG. 2 represents a best fit for an exponential curve which is very close to the linear fit. The linear fit in FIG. 2 corresponds to a current density of 50 nA/cm2, a dose rate of 9.2 kGy/sec, and an accumulated dose of 490 kGy. Since the 10 nA proton beam at 5.4 MeV and 40 nA/cm2 caused yellow discoloration/darkening at a delivered dose of 390 kGy, it is highly probable that the 50 nA/cm2 beam (490 kGy) also caused discoloration of the BoPEN film although it could not be seen since the area was subsequently ablated.

In contrast to the 1 nA fixed beam at 3 MeV, the subsequent 10 nA fixed beam suffered more than an order-of-magnitude larger, 18% decrease in its overall fluorescence in its first second of irradiation as compared to its initial signal, which must be due to immediate surface ablation. Similarly the 50 nA fixed beam suffered a 43% decrease in its overall fluorescence in its first second of irradiation as compared to its initial signal as seen in FIG. 5, and given its deep hole creation in just 33 seconds it can be considered a “fast” ablation.

FIG. 7 is a projection of the camera field-of-view for the digital image in FIG. 6A, taken at a working distance of 326 mm in accordance with embodiments. The camera was a Basler acA720-520 um with a 50 mm FL, f/1.4 lens. As shown in FIG. 6A, images were photographed through a chamber window of the 3.0 MeV proton beam irradiating the 191 μm thick BoPEN scintillator film in real-time, with the camera outside the vacuum chamber at an estimated working distance from the front of the camera lens to the scintillator film of ˜326 mm (as shown in FIG. 7).

As disclosed above, the very “first” digital image (1 ms shutter speed) taken within the 1st second of irradiation (i.e., at 2 fps and prior to significant ablation) at a beam current of 10 nA appears in FIG. 6A, which from the measured ablation area of 0.020 cm2 at the top surface of the hole yielded a current density of 500 nA/cm2 corresponding to a dose rate of 92 kGy/sec. The fitted beam profile for the image in FIG. 6A appears in FIG. 6B with a calculated a of 0.61 mm, and a FWHM of ˜1.22 mm, corresponding to a 97% full-bandwidth radius of 1.34 mm (i.e., 2.2σ) with an area of 5.6 mm2. From the previous experimental data of the rapid drop in average fluorescent signal (i.e., 18% decrease within the 1st second of irradiation), it's clear that at this dose rate ablation starts immediately upon beam exposure. However, from the progression of real-time camera fluorescence images over the 66 seconds of beam irradiation, it would appear that the ablation rate was relatively slow compared to that at the 50 nA beam current.

As disclosed above, an incident 5.4 MeV proton beam has adequate energy to pass through the 191 μm thick BoPEN film and exit with a residual energy of 3.26 MeV. However, at 3.0 MeV the proton beam only penetrates approximately 119 μm into the 191 μm thick BoPEN film. If the proton beam current density is sufficient to cause ablation and start “burning a hole” in the BoPEN film, then as the ablation proceeds the beam will penetrate further and further into the film, eventually exiting first at reduced energy and then almost at full energy once the hole has burrowed or punched through. Examination under a microscope confirmed that even at the 50 nA beam current, the ablated hole did not go all the way through the 191 μm thick film during the 33 seconds of beam irradiation, which followed the prior 66 seconds of much slower ablation at 10 nA.

The total estimated beam penetration depth was about 150-160 μm, and encompassed a maximum surface ablation area of ˜0.020 cm2, although the hole ellipsoid minor and major axes in the area of deepest penetration at the hole bottom was measured to be much smaller at about 0.4×0.6 mm (0.002 cm2). Based on the ablated hole area surface dimensions, the associated beam current density was 2500 nA/cm2 at 50 nA, corresponding to an accumulated dose of 15 MGy at a dose rate of 460 kGy/sec (see Table 1). At this dose rate, it is clear from the “average pixel signal” in FIG. 5, derived from each 1 ms photo/image at 2 fps, that ablation started immediately upon beam exposure (i.e., within a half-second). This result can be compared to the previous results for the 5.4 MeV proton beam at a current density of 2.4 nA/cm2, in which rad-damage occurred almost three orders-of-magnitude slower as it took 300 seconds for the bulk fluorescence intensity to decrease by 26%.

The ablated area/hole created by the 50 nA beam was elliptically shaped with measured minor and major axes of ˜1.4 mm×1.8 mm, corresponding to an equivalent circle with a radius of 0.80 mm and an area of 2.0 mm2. However the Gaussian fit distribution for FIG. 6A, as shown in FIG. 6B, corresponding to the 97% intensity full bandwidth has a beam radius of 2.2σ. This larger fluorescent emission area of 5.6 mm2 associated with the 2.2σ radius encompasses about 97% of the fluorescent signal area shown in FIG. 6A, and extends beyond the ablated hole. The fluorescent ellipsoid minor and major axis dimensions corresponding to the 2.2σ radius of 1.34 mm is 2.34 mm×3.02 mm, and corresponds to the estimated dimensions in FIG. 6A, with the camera image of the ellipsoid area containing 3,800 pixels. It follows from FIG. 7 that each pixel corresponds to a field-of-view image area of ˜38.2 μm×38.2 μm. The Basler acA720-520 um camera used for the FIG. 6A image has a 720×540 pixel CMOS sensor. It also follows that with the 50 mm focal length lens employed, the working distance (“WD”) from the front of the lens to the scintillator was about 326 mm, with the sensor field-of-view being 27 mm×21 mm as shown in FIG. 7.

The maximum beam current and minimum beam radius in the vacuum beamline pipe of a 250 MeV proton accelerator is typically ˜800 nA for a superconducting cyclotron with approximately a 1 mm beam radius. The associated beam current density is ˜25,000 nA/cm2. Under such conditions with a 25-50 μm thick BoPEN film scintillator, the dose rate could be 100-200 kGy/s, causing significant ablation of the BoPEN film and resulting in hole-burning within a minute or so. Good practice would dictate that the film radiation exposure in any one spot be limited to ten seconds or less.

For the above case of a 100-200 kGy/s dose rate, embodiments include a 5-way or 6-way-cross vacuum chamber that is designed to allow the BoPEN scintillator to be moved out of the beam within seconds after being moved into the beam to capture the required beam images. The proton beam image in FIG. 6A at a dose rate of 92 kGy/s provides an example of what such an image might look like. Although the BoPEN thickness in FIG. 6A is 191 μm, as compared to only 25-50 μm in the 5-way or 6-way-cross, the camera lens can be much closer to the scintillator in the cross than the 326 mm distance in FIGS. 6A, 6B and 7, so the solid collection angle is much greater to collect a larger fraction of the emitted photons from the thinner BoPEN film, and in addition a better light-sensitive camera could be employed than used in FIGS. 6A, 6B.

The low-energy proton beam tests at 3.0 MeV and 5.4 MeV for the 191 μm thick BoPEN film scintillator as summarized in Table 1 above covered a matrix spanning roughly three (3) orders-of-magnitude for the critical parameters of beam current density, absorbed dose and dose rate. The results of the described accelerated test program demonstrate the exceptional performance to be realized from the broad family of disclosed embodiments that have led to a wide variety of UFT (ultra-fast transmissive) high-resolution detection system embodiments for real-time monitoring of ionizing particle and photon beams. The targeted applications for the described embodiments below, include not only proton therapy, but all other types of particle and photon external beam radiation therapy (“EBRT”), as well as beam monitors for industrial and research accelerators including those used in nuclear and high energy physics, etc.

With regard to proton therapy, embodiments demonstrate an unexpected result that 5 minutes of testing at a beam particle energy of 5.4 MeV, a beam current density of 2.4 nA/cm2, and an irradiation dose rate of 200 Gy/s will not cause visual damage to the BoPEN scintillator, but would be roughly equivalent to the dose incurred in treating ˜30,000 patients assuming a conventional dose of 2 Gy per patient, or 3,000 patients at a FLASH dose of 20 Gy per patient. Thus radiation damage to a BoPEN film scintillator is not a significant issue and can be readily handled as disclosed below.

Given the previous estimate of 0.04% to 0.07% maximum accumulated scintillator radiation damage per week in a “typical” treatment room facility seeing 30 patients per day, embodiments have a need to advance a fresh area of scintillator film to the scintillator isocenter on a bi-weekly, monthly or possibly even quarterly basis; the latter period corresponding to a maximum estimated fluorescence loss of ˜0.9%. Therefore, as a practical matter it appears that having to measure the daily or weekly rad-damage contribution to scintillator non-uniformity can likely be ignored due to it being inconsequential, which has important implications. Specifically, calibration efforts in embodiments can be shifted to measuring and quantifying the other parameters that have to be monitored for achieving and maintaining an integrated system accuracy of 1% or better on a per patient daily basis. It follows that given the very small amount of rad-damage incurring on a weekly basis, a strategy of advancing the scintillator film, either by unwinding it from a spool (e.g., similar to advancing 35 mm film frame-by-frame in a camera) or by pushing a frame with the film mounted to it by a few centimeters on a periodic basis (e.g. weekly, biweekly, monthly, etc.) could be implemented via a variety of embodiments as disclosed below.

FIGS. 8A-C illustrate a system 800 that includes a two camera 840, single scintillator beam monitor in a light-tight enclosure employing a rolled scintillator spool configuration in accordance to embodiments. FIG. 8A is a perspective view with the top cover plate removed, FIG. 8B is a top view, and FIG. 8C is a section A-A view. The dotted arrows in all three figures show the direction of film movement from the feed roll to the take-up roll.

System 800 includes a two mirror 830, folded optical configuration which minimizes the light-tight enclosure depth/thickness while incorporating a mechanism for advancing the scintillator film 860 to minimize or eliminate having to correct for scintillator radiation damage. A relatively thick scintillator film such as 125-250 μm thick BoPEN film (i.e., 5-10 mils) is wound onto a small diameter (e.g., 2.5″) feeder spool 870 to an outer diameter (“OD”) that fits within the light-tight enclosure (e.g., ˜4″). This film could be of any width (e.g., 25-45 cm), and could contain a total length of about 20-25 meters of 191 μm BoPEN scintillator. In this embodiment, film 860 would be pulled across an active window area 812 onto a suitable take-up spool 872, and advanced by a stepper motor 880 that rotates the take-up spool spindle as required. An ultra-thin dark colored exit window 814, such as 15 to 25 μm thick black aluminum foil, is shown in FIG. 8C, while one of the UV-LED sources 850 and UV-photodiodes 852 are shown in FIG. 8A, with the two UV-LED/UV-photocell combinations 854 shown in FIG. 8B.

FIGS. 9A-D illustrate a system 900 that includes a two camera, single scintillator roll film beam monitor with linear translation of the scintillator spool system in a 6-way-cross vacuum chamber in accordance with embodiments. FIG. 9A is a cross-sectional view looking from the front with the scintillator film 940 positioned in the center of the beam path by linear position translators 950 and with cameras 902 and 904 in the top and bottom arms to achieve enhanced beam image resolution. The scintillator film is wound onto and stored on a small diameter feeder spool 930 and pulled across the beam axis transit area 970 (in FIG. 9B) onto a suitable take-up spool advanced by an internal (i.e. vacuum compatible) stepper motor 920 that rotates the take-up spool spindle as required. Also shown is a reducer nipple 990 that can connect to an external pressure bleed and/or vacuum line (not shown) to be used to break and then re-establish the beam monitor vacuum during system isolation for scintillator replacement (see FIG. 9C description below). FIG. 9B is the same cross-sectional view but with the scintillator film 940 translated vertically up and out of the beamline path region 970, by the linear position translators in their extended position 952. FIG. 9C is a perspective view of the closed system showing all 6 arms including the beam entrance and exit gate valves 912 and 910 that can be shut to isolate the beam monitor system and allow scintillator roll access and replacement without breaking beamline vacuum. FIG. 9D is a cross-sectional perspective view of FIG. 9C showing the ˜45° scintillator film angle with respect to both the beam angle of incidence and the viewing angle for both camera systems (also visible in FIG. 9A). It is noted that the 6-way-cross in FIGS. 9A-D is shown like all of the other 6-way-crosses with each arm at a 90° angle with respect to its nearest adjacent arm. However, to improve the photon collection angle/efficiency, one or both camera arms can be constructed at approximately a 45° angle with respect to the main body of the 6-way-cross housing the scintillator film so that the camera lens optical axis is at approximately a 90° angle with respect to the scintillator film plane.

FIGS. 10A-C illustrate a system 1000 that includes a roll film scintillator beam monitor in a smaller 6-way-cross vacuum chamber without linear translation capability in accordance with embodiments. FIG. 10A is a cross-sectional view from the front showing a camera 1004 and camera lens 1006 in the top arm and a PMT 1060 in the bottom arm; the latter for fast timing applications with enhanced light collection capability via a set of condensing lenses with the top lens 1050 located in the vacuum chamber just below the scintillator film 1040 and the bottom lens 1052 located just above the PMT 1060 in an ambient air environment. As in FIG. 9A, the scintillator film 1040 in FIG. 10A is at approximately a 45° angle with respect to the beam, camera and PMT. FIG. 10A shows the two UV-LED/UV-photodiode combination assemblies 1080 on opposite sides of the camera lens 1006. The scintillator film is wound onto and stored on a small diameter feeder spool 1030 and pulled across the beam axis transit region onto a suitable take-up spool 1024 advanced by an external stepper motor assembly 1020 that rotates the take-up spool spindle as required. FIG. 10B is a perspective view showing all 6 arms including the beam entrance 1001 and exit gate valves that allow system vacuum isolation and subsequent pressurization through the reducer nipple 1090 (in FIG. 10A) for scintillator roll replacement without breaking beamline vacuum. FIG. 10C is a close-up cross-sectional view showing the two UV-LEDs 1086 and 1088, and two UV-photodiodes 1082 and 1084 on opposite sides of the camera lens.

FIGS. 11A-D illustrate a system 1100 that includes a single scintillator-frame beam monitor in 6-way-cross vacuum chamber in accordance with embodiments. FIG. 11A is a cross-sectional view showing 4 of the 6 arms as seen from the front with a push-pull linear positioner on the left and a vacuum reducer nipple on the right. FIG. 11B is a perspective view showing all 6 arms of the closed system including a gate valve attached to the beam exit flange. FIG. 11C is a cross-sectional perspective view showing the tilted scintillator frame at approximately a 45° angle to the beam, camera and PMT. FIG. 11D is a close-up sectional view of the beam cross center showing a first condensing lens in the chamber vacuum region with the second condensing lens just below the viewport window in front of the PMT in an ambient air environment. Also just above the viewport UV window for the camera, on either side of the lens barrel are a pair of UV-LEDs and associated UV-photodiodes.

FIGS. 12A-C illustrate a system 1200 that includes a double scintillator-frame beam monitor in a 6-way-cross vacuum chamber in accordance with embodiments. FIG. 12A is a cross-sectional view showing 4 of the 6 arms as seen from the front, with a full-nipple and push-pull linear positioner added to each side as compared to only one side in FIGS. 11A-D. FIG. 12A shows one scintillator-frame on the left side with a second scintillator-frame mostly on the left side but covering the beam center. FIG. 12B is a cross-sectional view showing one scintillator-frame in each nipple with no scintillator in the beam center region. FIG. 12C is a perspective view of the closed 6-way-cross vacuum chamber. In system 1200, the scintillator-frame is at about a 45° angle with respect to the beam, camera and PMT.

FIGS. 13A-C illustrate a system 1300 that includes a double scintillator-frame beam monitor in a 6-way-cross load-lock vacuum chamber similar to FIGS. 12A-C, but with the addition of two gate valves, each positioned between the 6-way-cross body and the added reducer tees which have replaced the full-nipples in FIG. 12 in accordance with embodiments. The added gate valves convert this structure into a load-lock vacuum chamber, which allows scintillator replacement without breaking the system vacuum. FIG. 13A is a cross-sectional view (similar to FIG. 12A) showing 4 of the 6 arms as seen from the front. FIG. 13B is a cross-sectional perspective view that shows the approximately 45° scintillator-frame angle with respect to the beam, camera and PMT. FIG. 13C is a perspective view of the closed 6-way-cross load-lock vacuum chamber.

FIGS. 9-13 are based on “off-the-shelf” 6-way-cross configurations that have been modified such that the inner flanges associated with the two vertical tubes/arms as seen in FIGS. 9-13 are reduced or shortened to the minimum length required to weld each vertical flange to the cross body. The purpose of this modification is to position the cameras and/or PMTs as close as possible to the beamline axis/cross-center to improve the photon collection efficiency. However for some applications it is more important that the total length of the cross from the entrance-to-exit flange be minimized, and for such cases the two horizontal tubes/arms along the beam axis are shortened to the minimum stub size required to weld each flange to the cross body. For example, in the case of a beam monitor based on a 4″ O.D. tube system with 6″ diameter CF-flanges, the total length of the beam monitor including flanges from end-to-end can be reduced to under 6″. FIGS. 27A-B illustrate a system 2700 that includes both a side view (FIG. 27A) and perspective view (FIG. 27B) of the open central structure of the above reduced/shortened 4″ O.D. tube, 6-way-cross with 6″ diameter CF-flanges modified such that the total beam entrance-to-exit length is 5.9″ in accordance to vacuum chamber embodiments. The 4″ tubes that connect to the top and bottom 6″ CF-flanges that connect to the viewport windows, and subsequently to the full nipples that accommodate the camera and PMT, are also shortened such that the total end-to-end length for these two flanges is ≤7.9″ in accordance to embodiments. Depending upon the application requirements, the described priority can always be changed such that if the 4″ O.D. tube, 6-way-cross embodiment shown in FIGS. 27A-B is rotated by 90 degrees, then the end-to-end length for the two 6″ CF-flanges to the camera and PMT viewport windows would be ≤5.9″ while the minimum total beam entrance-to-exit length would be ≤7.9″ in accordance to embodiments.

FIGS. 14A-D illustrate a system 1400 that includes a two camera, two mirror, full-size single scintillator/window module beam monitor in a slim light-tight enclosure in accordance with embodiments. In one embodiment a “slim” light-tight enclosure is 5″ thick or less; however, depending upon the patient size requirements, scintillator dimensions, and image spatial and positional resolution specifications, the thickness can typically vary over a range from about 3″ to 7″. FIG. 14A is a perspective view of the components of a “drop-in” window/scintillator frame module. FIG. 14B shows how the window/scintillator frame module drops into one of the cover plate pockets. FIG. 14C is a perspective drawing of the two camera, single scintillator beam monitor enclosure with the top cover plate removed and positioned above the main structure. FIG. 14D is a cross-sectional view of the light-tight enclosure with drop-in ultra-thin window 1422 and window/scintillator 1460 modules, showing the folded optical design of camera, mirror and scintillator, and minimum scintillator field-of-view by camera-lens system on right side (i.e., within dotted line cone). Also shown in FIGS. 14C and 14D are a UV-LED source and UV-photodiode for internal calibration.

FIGS. 15A-C illustrate a system 1500 that includes a one camera, one mirror, half-size rectangular single scintillator beam monitor in a slim light-tight enclosure version of the embodiments shown in FIGS. 14C-D in accordance with embodiments. FIG. 15A is a perspective view assembly drawing showing the camera 1540, mirror 1530, ultra-thin window 1522, window/scintillator module 1560, UV-LED source 1550, UV-photodiode 1552, and the box construction with window cover plate 1520 and window/scintillator cover plate 1570 based on an internal frame structure. The actual enclosure shape and construction can vary and does not have to be rectangular (e.g. can be cylindrical). FIG. 15B is a cross-sectional view of the light-tight enclosure showing all of the basic described components. FIG. 15C is a perspective view of the enclosed system.

FIGS. 16A-C illustrate a system 1600 that includes a three camera version of the embodiments shown in FIGS. 15A-C in accordance with embodiments. The additional two side cameras do not have to be identical to the single top camera and can be selected for improved light-sensitivity, faster frame rates, and/or higher pixel resolution. FIG. 16A is a perspective view assembly drawing showing the three cameras 1640, 1644 and 1646, associated mirrors 1630, 1634 and 1636, ultra-thin window 1622, and window/scintillator module 1660, based on an internal frame structure. The actual enclosure shape and construction can vary and does not have to be rectangular (e.g. can be cylindrical). FIG. 16B is a cross-sectional view of the light-tight enclosure showing all of the basic described components. FIG. 16C is a perspective view of the enclosed system.

FIGS. 17A-B illustrate a system 1700 that includes a four camera version of the embodiments shown in FIGS. 14A-D for the full-size single scintillator-frame beam monitor with folded-optics in accordance with embodiments. The two additional cameras allow the field-of-view of each camera to be appropriately reduced to a scintillator quadrant, resulting most likely in selection of a different camera or different lens than in FIG. 14 for improved light-sensitivity, faster frame rates, and/or higher pixel resolution. FIG. 17A is a perspective view assembly drawing showing the four cameras 1740, 1741, 1742 and 1743, associated mirrors 1730, 1731, 1732 and 1733, ultra-thin window 1722, UV-LED sources 1750 and 1751, associated UV-photodiodes 1752 and 1753, and window/scintillator module 1760, based on an internal frame structure. The actual enclosure shape and construction can vary and does not have to be rectangular (e.g. can be cylindrical). FIG. 17B is a cross-sectional view of the light-tight enclosure showing all of the basic described components.

FIGS. 18A-B illustrate a system 1800 that includes a four camera, full-size double window/scintillator module beam monitor in a light-tight slim enclosure in accordance with embodiments. This embodiment is a double scintillator version of that shown in FIGS. 14A-D and incorporates both a front and back cover plate pocket design for the two “drop-in” window/scintillator frame modules. FIG. 18A is a perspective view assembly drawing showing the four cameras 1840, 1842, 1844 and 1846, with their associated mirrors including mirror 1836 coupled to camera 1846, aimed at the two window/scintillator modules 1860 and 1862. Cameras 1840 and 1842 through their respective mirrors are aimed at the bottom scintillator/window module 1860, whereas cameras 1844 and 1846 through their respective mirrors are aimed at the top scintillator/window module 1862. FIG. 18B is a cross-sectional view of the light-tight enclosure and like FIG. 18A shows two cameras with their respective folded-optics mirrors aimed at each scintillator.

FIGS. 19A-B illustrate a system 1900 that includes an eight camera, full-size double window/scintillator module beam monitor in a light-tight slim enclosure in accordance with embodiments. FIG. 19A is similar to FIG. 18A, but the number of cameras has been doubled, similar to FIGS. 17A-B compared to FIGS. 14A-D. Cameras 1940, 1941, 1942 and 1943, though their respective fold-optic mirrors, are each aimed at one quadrant of scintillator/window module 1960. Similarly, cameras 1944, 1945, 1946 and 1947, though their respective fold-optic mirrors, are each aimed at one quadrant of scintillator/window module 1962. Mirrors 1930 and 1931 for example are coupled to cameras 1940 and 1941. FIG. 19B is a cross-sectional view of the light-tight enclosure and like FIG. 19A shows four cameras with their respective folded-optics mirrors aimed at each scintillator.

FIGS. 20A-C illustrate a system 2000 that includes a four camera, single scintillator beam monitor employing a rolled scintillator spool configuration in accordance with embodiments, and is similar to the two camera version shown in FIGS. 8A-C. FIG. 20A is a perspective view, FIG. 20B is a top view, and FIG. 20C is a Section A-A view. FIGS. 20A and 20B show cameras 2040, 2014, 2042 and 2043, and their associated folded-mirrors such as 2030 and 2031. The dotted arrows in FIGS. 20A-C show the direction of film movement from the feed roll 2070 to the take-up roll 2072. In this embodiment, film 2060 would be pulled across an active window area 2012 onto a suitable take-up spool 2072, and advanced by a stepper motor 2080 that rotates the take-up spool spindle as required. An ultra-thin dark colored or black exit window 2014, such as 15 μm to 25 μm thick black aluminum foil, is shown FIG. 20C, while two UV-LED sources 2050 and 2051, and UV-photodiodes 2052 and 2053 are shown in FIG. 20A.

FIGS. 21A-B illustrate a system 2100 as perspective (FIG. 21A) and cross-sectional (FIG. 21B) views of a two camera 2140, full-size single scintillator-frame beam monitor in a light-tight box enclosure somewhat similar to that shown in FIGS. 14C-D, but using smaller size cameras (e.g., ˜1″×1″×1″) and not employing a folded optical system configuration with a mirror for each camera in accordance with embodiments. Each camera is thus aimed directly at the bottom scintillator plate 2130 resulting in the entire box enclosure being about 5 cm thicker than shown in FIGS. 14C-D. By removing the top cover plate and window 2110, and possibly even the side panels, the two cameras can be inserted just behind the exit nozzle or collimator (i.e., upstream) or alternately described as straddling behind the nozzle or collimator, and therefore integrated directly into the nozzle or collimator enclosure with the scintillator/window 2130 module inserted in the pocket of the exit cover plate 2120 located in front (i.e. downstream) of where the beam exits the nozzle or collimator. FIGS. 21A-B still incorporate one or two or more UV-LEDs and UV-photodiodes as found in all of the other beam monitors embodiments.

FIGS. 22A-C illustrate a system 2200 that is a four camera version of FIGS. 21A-B in accordance with embodiments. By removing the top cover plate and window as shown in FIGS. 22A-B, and possibly even the side panels, the four cameras 2240 can be inserted just behind the exit nozzle or collimator (i.e., upstream) or alternately described as straddling behind the nozzle or collimator, and therefore integrated directly into the nozzle or collimator enclosure with the scintillator/window 2230 module inserted in the pocket of the exit cover plate located in front (i.e. downstream) of where the beam exits the nozzle or collimator. FIGS. 22A-C still incorporate one or two or more UV-LEDs and UV-photodiodes as found in all of the other beam monitors embodiments.

In connection with the film rolls used in some embodiments, much longer rolls of the thickness and width disclosed above have been used for decades in aerial photography and advanced by motor drives at high speed—e.g., Kodak Aerial Ektacolor Print Film (SO-149) which with its color emulsion and gel backing has a total thickness of 213 μm. It is noted that standard 35 mm and 70 mm wide, motion picture film is typically advanced at 24 fps for “normal” motion but faster for slow motion, and some 70 mm IMAX films have been run at 48 fps (i.e., 200 meters/minute). If, for example, the BoPEN film were advanced 5 cm on a biweekly basis to shift the most likely rad-damaged scintillator center area (i.e., isocenter region) midway to the side, then the previously described 20-25 meter film length could last approximately 16 years. If the same BoPEN film were advanced 10 or 20 cm biweekly, then a single roll would last either 8 or 4 years respectively before requiring replacement.

Aerial films were previously made in four standard widths, 35 mm, 70 mm, 126 mm and 240 mm. These widths are the edge-to-edge dimensions and include sprockets on both sides, so for example the maximum image width on the 70 mm film is ˜58 mm, and on the 240 mm is 228 mm. The thinnest Kodak aerial film Estar “Ultra-Thin” Base made was 30 μm (i.e., 0.0012″) but was still strong enough to hold sprocket holes without tearing. However the standard Kodak Estar Ultra-Thin Base was 38 μm (i.e., 0.0015″), while the standard Kodak Estar Thick Base was 178 μm (i.e., 0.0070″). Film roll lengths for the Thick Estar Bases went from 100 to 800 feet, whereas film roll lengths of up to 2000 feet were standard for the other thinner Estar base films. A detailed thickness study by Kodak for their standard 240 mm wide Estar Base in a standard 30 meter length film roll yielded that “the thickness variation across essentially the entire roll length had a standard deviation of less than 1.85 μm”. However, within the 23 cm×23 cm aerial format picture area (i.e., 9″×9″) the standard thickness deviation was 1.0 μm. For Kodak 70 mm wide films, the spool core diameter was 31/32″ for film roll lengths up to 200 feet for the Estar Thin Base (64 μm), 150 feet for the standard Estar Base (102 μm), and 100 feet for the Estar Thick Base (i.e. 178 μm without emulsion and 184 μm thickness for B/W emulsions and 213 μm for their thickest color film). For longer rolls of 70 mm wide film, and for all lengths of 126 mm and 240 mm width film rolls, a spool core diameter of 2.125″ was used for all Estar film base thicknesses. Thus the suggested spool core diameter of 2.5″ disclosed above, and film length of 20-25 meters, are conservative given the standard specifications used for aerial films, as is the film thickness uniformity across the active area.

For the multi-arm cross, roll film embodiments, the 25 μm thick BoPEN should be ideal, especially considering that the BoPEN film is stronger than the Estar Base film (i.e., BoPET) used by Kodak, and the sprocket film holes that can tear in rapid advance photographic film systems are not required for the much slower advancing roll-to-roll embodiments described herein. In addition, the 12 μm thick BoPEN is also a potentially viable thickness for the roll-to-roll scintillator film designs such as the highly transmissive beamline vacuum cross monitors shown in FIGS. 9 and 10. In terms of the mechanical viability of such ultra-thin film rolls, it is noted that 12 μm thick linear low-density polyethylene (“LLDPE”) is available on a 3″ core in 18″ wide rolls of 1500 ft length (i.e., sold as a 47-gauge thick, polyethylene hybrid film), and even thinner 7 μm LLDPE film is available as 28-gauge film also on a 3″ core in 1500 ft length rolls. With regard to film strength, BoPEN is much stronger than LLDPE, having at least three times the tensile strength. As such, BoPEN rolls/coils are available in ultra-thin films down to 1.3 μm thickness in 12″ and wider size rolls, whereas 12 μm thick BoPEN film is available in 40″ wide rolls of 9800 ft length.

For embodiments that do not require roll-to-roll film advance systems, a variety of simpler yet more versatile transmissive beam monitor embodiments have been designed for fast exchange of different scintillator materials optimized for a wide variety of ionizing particles and energies including photons and neutrons in a wide range of film and sheet thicknesses (e.g., from ˜1 μm to ≥1 mm). For those monitors designed for beamline applications, both single-frame and double-frame, multi-arm cross structures are disclosed in which the scintillator is mounted to a stiff frame in contact with a push-pull mechanism as shown for three different 6-way-cross embodiments in FIGS. 11A-D, and 12A-C, and 13A-C below. If only one camera is required and no PMT, then a less expensive 5-way-cross can be employed, and depending upon the frame and arm (or nipple) length, the scintillator can be pulled out of the beam path entirely and pushed into the beam path only when beam monitoring is required (e.g., as shown in FIG. 12B). Another advantage of mounting the scintillator film to a rigid frame, as compared to a roll-to-roll system, is that thick scintillators (e.g., 0.5 mm) cannot be rolled onto a small diameter spool, while the thinnest scintillator films of only a few microns cannot be reliably advanced across the beam axis transit area without risk of damage when pulling it from the feed spool 1030 onto the take-up spool 1024 as in FIG. 10A. As disclosed, 3, 6, 12 and 191 μm thick BoPEN films have unexpected characteristics of radiation hardness and fast recovery (see FIG. 3). Other embodiments can use 1.3 μm BoPEN films or 250 μm thick BoPEN films for several different applications.

All of the embodiments disclosed herein with associated figures/drawings incorporate the previously described system/hardware required for internal calibration and beam image analysis. The calibration system and its operation can be initiated either manually or automatically (e.g., on a pre-programmed schedule) and is based on activating an internal UV-LED source or sources to illuminate the scintillator film for a short time (e.g., seconds) and capturing images of the fluorescence intensity pattern and comparing them to previous images by means of an appropriate computer system to detect any changes in the system response, including changes in the scintillator fluorescence or camera sensor such as might be caused by radiation damage, etc. In order to monitor the stability of the UV-source, each UV-LED is itself monitored by a dedicated proximity UV-photosensor such as a photodiode to correct for any source intensity change or drift over time. The computer system in one embodiment is a dedicated, low-latency, fast-PC (personal computer) or workstation, etc., having a processor that executes instructions. In other embodiments, the computer system is a customized FPGA based PCB (printed circuit board) or frame grabber, although more likely a frame grabber connected to a computer. For some systems the FPGA could be partially or fully embedded in the camera(s). The computer system besides performing internal calibration checks, is also programmed to perform image analysis in real-time of the beam as it irradiates the scintillator so as to monitor and analyze in two-dimensions (“2D”) the beam position and beam shape, beam movement, the beam intensity profile including tail, beam fluence and external dosimetry, and beam angular divergence in the case of the beam monitor configuration incorporating two or more scintillators in the beam path and separated by an appropriate distance. In addition, because all of the embodiments incorporate one or more machine vision cameras oriented at an angle to the scintillator plane, all of the camera images will incur perspective/tilt distortion (i.e., keystoning), while the camera lenses, especially due to their close working distance, will exhibit some amount of optical distortion as well as vignetting, and the camera sensors themselves can never be perfectly uniform in terms of each pixel having exactly the same response. All of these system hardware related non-uniformities can be corrected by calibration of the integrated system and frequently checking this calibration by taking repeated images of the system response to the UV-source illuminated scintillator and automatically adjusting the calibration as needed by the computer system.

In order to minimize maintenance and down time, and to further optimize the design, fast scintillator replacement is required in embodiments. This is achieved in all of the embodiments shown in FIGS. 14-19 and 21-22 by the design of a thin (e.g., 2 mm to 3 mm thick), replaceable large window/scintillator frame module assembly such as 1460 in FIG. 14A that can be easily accessed and dropped from the outside into a small pocket (˜2-3 mm deep) in the front and/or back enclosure cover plate(s) 1414, and secured from the outside with a thin 1-2 mm thick retaining frame 1410 as shown in FIG. 14B. The window/scintillator module assembly 1460 consists of ultra-thin window 1402, such as 15 μm to 25 μm thick black aluminum foil, and a scintillator film or sheet 1406, with both components attached or glued to opposite sides of a thin frame 1404, Replacement of the ultra-thin window/scintillator module using the design in FIGS. 14A-D should take only a few minutes. If only one scintillator/window module 1460 is employed, then an ultra-thin window by itself such as 1402 is glued to the frame 1404 without adding the bottom scintillator plate. The window module itself, without a scintillator component, is shown as 1422 in FIGS. 14C-D and fits into the pocket of cover plate 1420 and held in place by a retaining frame (e.g., 1410 in FIG. 14B) as shown in FIG. 14C. FIGS. 14C-D show one such embodiment based on a 2 camera, 1 scintillator arrangement, with both cameras 1440 and 1442 indirectly aimed at the back scintillator 1406 in the back window/scintillator module 1460 through their respective folded-optics mirrors 1430 and 1432. Also shown are UV-LEDs 1450 and UV-photodiodes 1452 in FIGS. 14C-D.

Embodiments include a number of different light-tight enclosure beam monitors incorporating one or two scintillators and from one (1) to twelve (12) or more cameras, depending upon the desired beam spatial/positional resolution and the required scintillator active area size which for EBRT applications can typically extend up to about 40 cm×40 cm. In general, for a 20 cm×20 cm scintillator, the intrinsic 2D position resolution should be on the order of ˜0.03 to 0.2 mm, depending upon the required UFT beam monitor specifications. However, no matter how many cameras are employed, such as 1, 2, 3, 4, 6, 8, 10, 12 or more (see FIGS. 8 and 14-22), the software required for stitching multiple camera images together is commercially available for scientific, industrial, medical, consumer applications, etc. For example, a number of smartphones now employ multi-camera systems, such as the Samsung Galaxy S10 and S10+, which use 3 cameras to stitch together high quality images with minimal distortion that can cover the full range from ultra-wide angle to telephoto. The above disclosed platform can perform the equivalent of stitching together multiple images as they stream in from multiple machine vision cameras in order to track and analyze the moving particle beam or photon beam as it travels both horizontally and vertically across the scintillator surface. For such streaming images the software in embodiments is mostly FPGA based, on a multi-camera frame grabber based system that can provide calibration and corrections for optical distortions and equipment/system non-uniformity.

Depending upon the application requirements in terms of: image/pixel resolution, low-light sensitivity, pixel bit depth (i.e., gray scale), exposure time (i.e., shutter speed), frame rate, and image processing speed including system latency, camera images can be streamed live, processed and analyzed in real-time at rates potentially as fast as 25-100 μs per image (i.e., 10,000-40,000 fps) depending upon the system hardware, firmware and software, including the choice of camera interface. For example, machine vision cameras operating at over 30,000 fps, corresponding to a timing resolution of ˜33 μs, can still provide sub-mm image resolution for embodiments such as the above multi-camera, 20 cm×20 cm, or even 40 cm×40 cm, scintillator EBRT beam monitors at a cost of about $5K to $8K per camera is single unit quantities. The larger size 40 cm×40 cm scintillator beam monitoring systems in accordance to embodiments employ 4 or more cameras, and are configured if desired with two different types of cameras—for example in a single-scintillator 6-camera configuration there could be four relatively inexpensive high picture resolution, low frame rate (fps) cameras, plus two of the more expensive high fps cameras; other combinations are also possible such as 4 slow and 4 fast cameras, or 4 low sensitivity and 4 high sensitivity cameras in a 8-camera system. In fact, low cost, high spatial resolution, low sensitivity, low fps cameras could even be paired side-by-side with ultra-fast, high sensitivity, ultra-compact PMTs (e.g., Hamamatsu H11934 series with dimensions of 30 mm×30 mm×32 mm) with camera lenses coupled to each PMT thereby viewing the same scintillator area as the camera. The PMTs would provide the low-light sensitivity and dose rate information with ultra-fast ns and sub-ns response capability (e.g., 10 ns is equivalent to 100,000,000 fps). For applications requiring frame rates of 1,000-2,000 fps or slower, smaller size machine vision cameras can be procured for $1K (see below).

FIGS. 23A-B illustrate camera images through a vacuum chamber window of a ˜3.6 mm diameter proton beam, moving at 80 mm/ms, irradiating a 191 μm thick BoPEN scintillator, with a 10 μs exposure in accordance to embodiments. The camera used is a Basler daA1280-54 um with a 25 mm FL, f/1.4 lens, at a working distance of ˜350 mm, with a pixel field-of-view of 48 μm×48 μm. FIG. 23A constitutes an image of the camera's full field-of-view.

In FIGS. 23A-B, the proton beam energy was 5.4 MeV at a 10 nA beam current. FIG. 23B is an enlarged and cropped image with the background digitally removed of the beam spot area in FIG. 23A, showing the pixel resolution detail including the intensity distribution and beam shape and dimensions which covers an irregularly shaped elliptical area of ˜60×100 pixels. The beam horizontal “smear” during the 10 μs exposure due to the 80 mm/ms movement is only ˜0.8 mm, or about a 22% elongation.

FIG. 24 illustrates a 1 ms exposure of a captured image using the same camera/lens as in FIG. 23, but of a ˜2 mm diameter proton beam irradiating an ultra-thin 12.2 μm BoPEN film while moving back and forth in a rastered zig zag pattern at 40 mm/ms in accordance to embodiments. As with FIGS. 23A-B, the proton beam energy is 5.4 MeV at a 10 nA beam current, but at a lens working distance of ˜390 mm, corresponding to a somewhat larger 55 μm×55 μm field-of-view pixel resolution. Similar images have been captured on BoPEN films as thin as 3.0 μm, with plans to irradiate a 1.3 μm thick BoPEN film in the near future.

One embodiment uses Universal Serial Bus (“USB”) as the system interface, hardware and processing software, which is capable of processing and analyzing images at rates up to about 1,000-2,000 fps (i.e., 1.0 ms to 0.5 ms). For higher performance, embodiments use a number of faster camera interfaces for interfacing with high-speed FPGA based frame grabber hardware, firmware and software, to process and analyze the streaming images at much higher speeds, including CoaXPress 2.0 (CPX-12), GigE (10 Gigabit Ethernet), Camera Link HS, etc.

Referring again to FIGS. 11A-D, both cross-sectional and perspective views of a single scintillator-frame beam monitor are shown. FIG. 11A shows 4 of the 6 arms of a modified CF-flange 6-way-cross vacuum chamber configuration, although any type of flange system can be used (e.g., ConFlat, KF/QF, ISO-K, ISO-F, ASA, Wire-Seal, etc.). The two arms not shown in FIG. 11A are perpendicular to the plane of the drawing where the beam enters 1101 the cross center as seen in FIG. 11B. Either or both of these two arms can incorporate an optional gate valve attached to one or both flanges for vacuum isolation. FIG. 11B shows one such gate valve 1110 attached to the exit arm flange. As the integrated exposure of the scintillator to the ionizing beam accumulates over time, so does the radiation dose which would typically be concentrated in or near the beam pipe center. Therefore, at such time that the scintillator radiation damage becomes significant, the scintillator-frame unit 1140 is nudged or pushed by shaft 1172 an appropriate distance (e.g., ˜1 cm, or more) on its track 1145 in FIG. 11D towards the opposite side (i.e., the right side in FIGS. 11A-D) to bring unexposed or minimally exposed scintillator film into the central beam path region. This linear shift/movement can be accomplished either manually, or controlled pneumatically, or by a stepper motor as indicated by the linear positioner 1170 shown on the left side in FIG. 11A.

FIG. 11A shows the camera 1104 and camera lens 1106 in the top nipple, along with the UV-LED/UV-photodiode assembly combination 1180, and a conical reducer nipple 1190 to the vacuum-exhaust/air-bleed line (not shown). FIG. 11D shows a close-up of the two small UV-LEDs 1186 and UV-photodiodes 1182 positioned on opposite sides of the camera lens. The PMT 1160 in the bottom nipple is shown most clearly in FIG. 11C, while the two condensing lenses 1150 and 1152 on either side of the viewport window 1156 in the bottom nipple are best seen in FIG. 11D. FIG. 11D provides a close-up magnified view of the 6-way-cross center area in which the two viewport windows 1155 and 1156, UV-LEDs, UV-photodiodes, two scintillator-frame tracks, and the two condensing lenses are most easily seen.

All of the embodiments disclosed herein include at least one UV illumination source, with at least one UV photosensor to monitor the stability of each UV source. The UV source employed in embodiments for the BoPEN scintillator is a UV-LED with peak emission at ˜280 nm, where the BoPEN scintillator film essentially absorbs at least 99% of the source photons at the film surface within a ˜0.1 μm thick layer. The UV photosensor used to monitor the UV-LED in embodiments is a UV-photodiode. If needed, the UV source and/or UV photosensor can be coupled to a suitable UV bandpass or UV shortpass filter. When the rad-damage in any particular area starts to become significant, the scintillator-frame is pushed slightly towards the far side until such time as the frame has been pushed completely to the far side as shown in FIGS. 11A and 11D. Once the scintillator has been fully radiation damaged along its useable length, the scintillator-frame is then pulled back to its initial position and the scintillator-frame replaced. Replacement requires breaking vacuum in the 6-way-cross chamber, but because of the excellent BoPEN rad-hardness (see Table 1 above) it might be possible to schedule such replacement during preplanned downtime periods allocated for general maintenance.

If the scintillator 6-way-cross chamber includes both entrance and exit gate valves, then breaking vacuum is limited to the small chamber volume with no impact on the rest of the beamline and so scintillator-frame replacement can be done whenever convenient and should only take about an hour or so including ambient pressurization and re-evacuation. Other features of consequence are the machine vision camera in the top arm, the PMT in the bottom arm, the push-pull linear positioner on the left side, and the reducer nipple on the right side which is connected to a small vacuum pump system (not shown) with a bleed valve for chamber pressurization followed by re-evacuation. Also not shown are the described beam entrance gate valve, although the exit gate valve is easily seen in FIGS. 11B and 11C and so an entrance gate would look the same when attached to the entrance flange in FIG. 11B. Both the camera and PMT in their respective nipples are kept in an air atmosphere at ambient pressure. This is achieved on the camera side by having a UV transparent viewport window 1155 inserted between the cross flange and the camera nipple. The reason for a UV window is because the UV-LED in this embodiment is located on the side of the camera lens. If the camera power consumption is large enough to cause significant heating, then the nipple back flange can be left open or vented to facilitate air cooling by either natural or forced convection as long as the front of the camera or lens barrel is appropriately light shielded behind the UV-LED and UV-photodiode. The same viewport window arrangement is used on the PMT side, but for the BoPEN scintillator or other scintillators with shorter decay times (e.g., EJ-200, EJ-204, EJ-212, EJ-228, EJ-262, etc. from Eljen Technology) with emission peaks in the violet-blue-cyan region, the viewport window 1156 can be glass.

To maximize the PMT light collection efficiency a set of highly efficient, high transmission glass (e.g., Schott B270) aspheric condensing lenses are employed with an f/number that can be less than 1.0 (e.g., between f/0.6 to f/0.9). For maximum efficiency, the first condensing lens 1150 is located inside the cross vacuum chamber just below the scintillator/frame, while the second lens 1152 is located just below the glass viewport window 1156 and in front of the PMT at ambient pressure as shown in FIG. 11D. Both lenses can be anti-reflection coated for maximum light transmission and the second lens located in front of the PMT can further reduce reflection loss by optically coupling it to a matching refractive index plastic or glass light guide (e.g., cylinder) thereby eliminating the air gap completely. The PMTs should be selected for minimum jitter (e.g., ≤0.3 ns), maximum quantum efficiency (e.g., ≥22%), and most importantly for maximum gain (e.g. >1×106). Besides having a short decay time, the scintillator should have a high light yield and if capable of total internal reflection (TIR) could have a reflective coating deposited on the non-collecting surface, or surface roughened to eliminate TIR on the light collect surface, or for optimum TOF performance could employ two matching PMTs with two sets of condensing lenses in the 6-way-cross (i.e., replacing the camera with a second PMT).

The embodiment shown in FIGS. 12A-C is similar to that in FIGS. 11A-D, but with the addition of two horizontal full-nipples 1290 and 1292 to accommodate a dual scintillator-frame configuration. Similarly the embodiment in FIGS. 13A-C is quite similar to that in FIGS. 12A-C, but with the important addition of two vertical gate valves 1310 and 1311 that effectively transform the embodiment in FIG. 12 into the load-lock vacuum chamber of FIG. 13. FIGS. 12A-B show 4 of the 6 arms of the customized 6-way-cross vacuum chamber; the two arms not shown are perpendicular to the plane of the drawing where the beam enters and exits the cross center. FIG. 12C is a perspective view showing all 6 sides/arms, including the two perpendicular arms where the beam enters 1201 and exits 1202 and which can incorporate one or two optional gate valves such as 1310 and 1311 as shown in FIGS. 13A-C and previously discussed for FIG. 11. The dual scintillator-frame embodiments employ either a straight track 1245 or a segmented track 1345, as shown respectively in FIGS. 12A and 13A that goes through all three chamber sections on which the scintillator-frames can be pushed or pulled. If two identical scintillators are employed, the maximum time before scintillator replacement can be essentially doubled. The dual scintillators 1240 and 1241 in their frames as illustrated in FIGS. 12A-B also allow two different scintillator materials to be employed, each selected for a different purpose. For example, one scintillator might be selected for minimum film thickness and maximum beam transmission (e.g., BoPEN), with the other selected for minimum decay time and rise time to provide the fastest possible timing when coupled to an efficient light-collection system such as the condenser lens system shown in FIGS. 12A-B, which can be seen more clearly in FIG. 11D as lens elements 1150 and 1152, and a fast PMT 1060, 1160 & 1260 in FIGS. 10A, 11C and 12B respectively for sub-ns TOF (time-of-flight) measurements. With embodiments, timing resolutions of ≤0.1 ns are achievable for highly ionized, high-Z (i.e., atomic number) beams using the 6-way-cross beam monitors shown in FIGS. 10A, 11A-D and 12A-B.

As discussed above, for the embodiment shown in FIGS. 12A-C, the two scintillators mounted in their respective frames can either be identical or the first scintillator-frame combination 1240 might be selected for fast timing (e.g., BC-400 from Saint-Gobain) and the second being a thinner scintillator 1241 of different composition, such as BoPEN, selected for maximum beam transmissivity with minimal beam scattering and energy loss (i.e., from an incident photon or particle beam such as protons, ions, electrons, neutrons, etc.). The scintillator-frame 1241 in its initial start position is shown in FIG. 12A before being nudged or pulled in small steps towards the opposite (i.e. right) side as in FIG. 12B to bring unexposed or minimally exposed scintillator film into the central beam path region. Such linear movement can be accomplished either manually with linear push-pull positioners 1220 and 1230 in FIG. 12A, and 1320 and 1330 in FIG. 13A, or controlled pneumatically or by a stepper motor.

As the integrated exposure of the scintillator to the ionizing beam accumulates over time, so does the radiation dose which would typically be concentrated in or near the beamline cross center 1250 shown in FIG. 12B. The two horizontal length nipples 1290 and 1292 hold the two scintillator-frames 1240 and 1241 in FIG. 12B with scintillator-frame push-pull linear positioners 1220 and 1230 attached to their respective nipple and scintillator-frame that push or pull the two scintillator frames on their tracks from the left side across the beam center area 1250. When fully “used up” (i.e. radiation damaged) the center scintillator-frame is pulled from the right into the right side nipple chamber 1292 in FIG. 12A (or 1392 in FIG. 13A) for removal, while the left scintillator-frame 1340 in the left side nipple 1390 in FIG. 13A can be pushed into the center of the 6-way-cross where the beam enters through flange 1301 in FIG. 13C. The top vertical nipple contains the camera 1004, 1104, 1204, 1304 in FIGS. 10-13 respectively, and camera lens 1006 or 1106, while the bottom vertical nipple contains the PMT 1060, 1160, 1260 or 1360 (or SSPM). The two vertical nipples containing the camera and PMT are at ambient pressure and isolated from the vacuum by their hermetically-sealed windows—e.g. 1155 and 1156 shown in FIG. 11D. The scintillators are pushed-pulled along a three section channel/rail or track 1345 in FIGS. 13A and 13B with two breaks or open-segments of ˜2 cm each through which the two gate valves 1310 and 1311 can close. The scintillator-frames 1340 and 1341 can each be removed without breaking vacuum by closing a gate valve. With the gate valves closed, each scintillator nipple section can be individually pressurized for scintillator replacement and then re-evacuated using a small pump through the two nipple tee sections 1391 and 1393.

The embodiment shown in FIG. 12 does not have any gate valves to isolate each nipple during scintillator replacement, so “nipple” 1290 is actually a reducer tee with reducer flange 1291 for connection to an external pressurization line and optional vacuum line to minimize downtime during scintillator replacement. This arrangement is similar in function to the conical reducer nipples 1090 and 1190 shown in FIGS. 10A and 11A for attachment to an external pressurization/vacuum-exhaust line. Although not easily seen in FIG. 12 or 13, for internal calibration purposes the described embodiments include UV-photodiodes 1082 and 1084, as shown on each side of the camera lens 1006 in FIGS. 10A and 10C, to monitor the output of each UV-LED 1086 and 1088 in order to correct for changes in the UV-LED output luminosity. This internal UV-LED/UV-photodiode calibration system 1080 in FIG. 10A, also shown as 1180 in FIG. 11A, and 1182 and 1186 in FIG. 11D, is also be used to monitor and correct for any changes with time or temperature of the camera sensor output.

Depending upon the specific application and beamline monitoring requirements, a number of variations of the 6-way-cross structures described above and in FIGS. 9-13 are available. For example, if a PMT is not required, then a 5-way-cross can be used, but if spatial resolution, sensitivity and accuracy are paramount, then the 6-way-cross could be used with two cameras—i.e. the second camera replacing the PMT as shown in FIG. 9A. Alternatively, if the beamline monitor is to be optimized for time-of-flight (“TOF”) measurements with the highest timing resolution and accuracy required, then two closely-matched PMTs with two sets of condensing lenses can be employed, and the camera eliminated as discussed previously. For other applications only a 3-way-tee or 3-way-wye might be required, or a 4-way-cross could be utilized.

Not all beamline monitoring systems need to be integrated into a vacuum beam pipe environment, including segments of electron and neutron beam delivery systems. Such systems, however, can still utilize the various multi-arm-cross embodiments disclosed herein. For monitoring the beam in air, the crosses do not have to be evacuated but can simply be made light-tight by adding a thin foil or dark/black polymer film window, or some polymer-foil combination thereof, to the entrance and exit flanges. For enhanced scintillator recovery, the air atmosphere can be replaced by any gaseous atmosphere including oxygen or oxygen enhanced mixtures, or pure nitrogen or argon or any other type of specified atmosphere.

All of the embodiments with cameras include the camera or cameras viewing the scintillator at various angles of incidence or reflection, the latter indirectly via a folded-optics mirror system. Parameter optimization determines the most appropriate camera lens angle of incidence with respect to the normal to the scintillator plane (i.e. surface) or mirror for each application. For most of the embodiments disclosed here, the camera lens viewing angle with respect to the scintillator will typically fall within the range of 25-65°, with an average value of ˜45°. For the camera images captured in FIGS. 6, 23 and 24, the camera angle of incidence with respect to the scintillator normal typically fell within 20-30°. For the embodiments in FIGS. 8-22, the mean camera angle with respect in the scintillator normal, or mirror normal in the case of FIGS. 8 and 14-20 was typically 40-50° but can be increased to minimize the enclosure depth or thickness. However, any camera angle greater than a few degrees will create some angular distortion of the image, and depending upon the amount of distortion, a circle, for example, can look like or appear as a distorted ellipse. In fact, for a camera to scintillator angle of just 5°, the distortion will start to be noticeable, and at a 10° angle the distortion will definitely be noticeable. Therefore at the 20-30° angles for the images in FIGS. 6, 23 and 24, the discussed ellipsoids might actually be circles but only appear to be ellipsoidal due to this distortion. In the case of images of a moving or rastering beam, the beam motion will further distort the shape of the image in the propagation direction (see FIGS. 23 and 24). These image distortions can be corrected by software.

FIG. 25 illustrates a four plate light baffle 2500 for air circulation within a light-tight enclosure by natural convection in accordance to embodiments. A more efficient light-tight air circulation arrangement by means of forced convection can be realized by the addition of one or more miniature fans.

FIG. 26 is a photograph of a 2×2 inch rectilinear image taken at 45° tilt angle in accordance to embodiments. FIG. 26 shows perspective distortion, also known as the keystone effect (e.g., image foreshortening caused by the angle-of-tilt with respect to the lens orientation).

The angular distortion disclosed above caused by the angle of tilt, as shown in FIG. 26, is known as a perspective distortion, but also called tilt distortion, the keystone effect, keystone distortion, or simply keystoning. A familiar example occurs when taking a picture of a tall building from the ground, with the building looking more and more trapezoidal the taller it is and the greater the camera angle of tilt. None of the images presented herein have been corrected for this distortion, but it is easily corrected in real-time with modern image editing software. Obviously the greater the angle of camera lens tilt, the greater the distortion, and the greater the difference in image resolution at the image top edge as compared to the bottom edge. For example, in FIG. 8 and FIGS. 14-22, at an average camera lens viewing angle with respect to the scintillator of ˜60° instead of the 45° angle in FIG. 26, there will be approximately a factor of two (i.e., 2×) difference in image resolution at the image top edge compared to the bottom edge for a camera focused on the center of a 10 cm×10 cm scintillator quadrant (e.g., FIG. 17 and FIGS. 19-20), with a 10 cm working distance from the camera lens to the closest point of the image field (i.e., top of quadrant). Angles of 60° or even greater are necessary for achieving the thinnest beam monitor configurations possible, such as those required by BNCT, disclosed below.

Although the above embodiments have mostly been described and tested in terms of their applicability to proton beams and proton beam therapy, these embodiments are applicable to all types of particle beams including those for particle beam therapy (e.g., protons, helium-ions, carbon-ions, electrons, etc.), as well as neutron particle beams. Fast neutrons can still benefit from the described advantages associated with BoPEN scintillator films, but slow to thermal neutrons require boron or lithium or gadolinium doped scintillators, such as boron doped EJ-254, which is of interest for boron neutron capture therapy (“BNCT”) and gadolinium neutron capture therapy (“GdNCT”). Although most of the disclosed embodiments have referenced the BoPEN scintillator, none of the embodiments are scintillator specific, so any scintillator material can be employed. The described embodiments are also of interest to particle research accelerators. The particle beams used for such research include everything from electron and muon beams, to rare isotope and exotic heavy-ion and radioactive ion beams such as highly charged uranium ions beams (e.g., U-238 with a net charge greater than +60). In addition there are tens of thousands of particle beams used by industry and various versions of these embodiments could find application there.

Embodiments can also be used for external beam radiation therapy (“EBRT”) based on high energy photon beams (e.g., MeV gammas and/or X-rays). Embodiments disclosed herein, such as those in FIGS. 8 and 14-22, have advantages over the known ionization chamber beam monitors that find wide application in photon EBRT, with even more advantages for FLASH therapy. These advantages over ionization chambers include up to two orders-of-magnitude faster beam profile imaging time (e.g., ˜10 μs vs. 1000 μs), at least one order-of-magnitude better intrinsic 2D position resolution (e.g., ˜0.03 mm vs. 1 mm), and more than one order-of-magnitude higher dose rate capability (e.g., ˜5,000 Gy/s vs. 120 Gy/s). Although all of the light-tight ambient pressure beam monitor enclosures shown in FIGS. 8 and 14-22 are rectangular in shape, this is not a requirement or a limitation, and thus other shaped enclosures can be employed such as cylindrical shaped beam monitor enclosures. In addition, the thickness of the above referenced ionization chambers and the scintillator based UFT beam monitor can be almost the same, depending upon specifications.

The highest performance cameras with the largest sensor size, best low-light sensitivity, highest bit depth (i.e., pixel gray scale range), highest frame rates, most sophisticated embedded FPGA circuitry, and thus the highest data transmission output in terms of MB/s, consume the most power. The maximum power consumption for such cameras could be on the order of ˜10 watts per camera, although the standby power when the camera isn't running would likely be much less depending upon the camera. And for smaller cameras, such as those used for the images in FIGS. 23 and 24, the average power consumption was only 1 watt. Nevertheless, for the case of higher power consumption cameras operating in a sealed enclosure, heat generation followed by heat build-up could potentially be a problem if not adequately addressed.

Several solutions exist to the potential problem of heat build-up, including the use of a series of internal baffled air-vents with staggered holes to block light leaks, as for example in FIG. 25, that would allow cooling by natural air convection. The total thickness of such vents need not be more than 2-3 cm. If necessary this concept can be augmented by forced convection if coupled to one or more miniature fans (e.g., 2-inch to 4-inch blade diameter) for forced-air cooling. A minimum of two such vented light baffles would be required, preferably attached to opposite sides of the scintillator box enclosure—i.e., an entrance baffle for air/oxygen in-flow, and exit baffle for air/oxygen out-flow. The FIG. 25 (Side View) drawing uses dashed arrows to illustrate the air flow through the staggered holes of the light baffle vent from entrance 2501 to exit 2502. The purpose of the light baffle is to facilitate continuous circulation and exchange of cool ambient air flow through the scintillator box enclosure, or of cold gas such as cryogenically cooled nitrogen, or of oxygen enriched air, or even pure oxygen circulation through the scintillator box enclosure, while preventing or minimizing light leakage. The motivation for oxygen circulation in the scintillator box enclosure is that oxygen diffusion into the scintillator can potentially minimize scintillator radiation damage by facilitating partial recovery or repair of scintillator damage by oxygen scavenging of radiation damaging free-radicals created in the scintillator by the incident ionizing radiation beam.

Alternatively the light-tight enclosures can be sealed around the camera lenses, with the camera body protruding out of the light-tight enclosures and thereby venting the camera heat to the external ambient open-air environment. For the 6-way-cross systems, a custom short nipple can be made with a light-tight seal (e.g., double O-ring) to the camera lens, thus leaving the camera body protruding outside and beyond the nipple flange to the external atmosphere. There is also the option of using active cooling of the camera or silicon image sensor, or even cryogenic cooling, as some cameras are sold with thermoelectric cooled sensors. Finally, each camera and/or sensor could be calibrated for their signal response or drift as a function of temperature, and then the temperature of the camera or sensor in its enclosure monitored and its signal response automatically corrected by software.

For the detection of neutrons in EBRT applications, the two most obvious locations for a neutron beam monitoring system might be: (1) immediately after the Li target, but before the moderator, where mostly slow neutrons but perhaps some fast neutrons (e.g. ˜0.8 to 1 MeV) are typically generated by a ˜2.6 MeV proton beam, and (2) at a location after the moderator where the neutron energy is degraded for many boron neutron capture therapy (BNCT) treatment regimens to the epithermal energy range but more broadly across the range from thermal to slow or even fast neutrons. If only one neutron beam monitor is to be employed, the most important location would be right after the moderator and in front of the patient. Recent trials in Finland suggest that 1-30 keV “slow” neutrons constitute a practical energy range for BNCT treatment. Essentially all of the beam monitor embodiments disclosed herein should work well for the detection and monitoring of neutrons created immediately after the Li target in location (1) above, where the neutron intensity of the BNCT machine is estimated to be on the order of ˜1013 n/s, corresponding to about 5×1013 scintillating photons per second from a 0.2 mm thick BoPEN film. Therefore a much thinner scintillator film can to be used to minimize interaction with the beam, and still produce a huge amount of scintillating light. For example, a 12 μm thick BoPEN film scintillator should yield about 3×1012 scintillating photons per second. However, for beam monitoring in location (2), the added moderator plus energy filtering greatly reduces the number of epithermal neutrons by at least several orders-of-magnitude, which are significantly more difficult to detect anyway due to their lower energy than the more energetic “slow” neutrons in location (1). This means that for neutron beam monitoring after the moderator, B10 or another high neutron cross-section isotope (e.g., Li6 or Gd) loaded scintillator is required to increase the deposited energy in the scintillating host. Such scintillators are available in plastic sheets and can be incorporated in the scintillator-frame embodiments disclosed above and shown in FIGS. 11-22.

For BNCT head and neck EBRT therapy, the patient's head is typically positioned very close to the neutron beam exit nozzle, and therefore the thinnest profile beam monitors are required corresponding to the largest camera-lens angles with respect to the scintillator normal (e.g., 60°-70°). Modified versions of FIGS. 14-19 with camera angles of ≥60° have been designed for such applications with total beam monitor thicknesses of ˜6 cm to 8 cm (i.e., from entrance to exit window), which is almost the same thickness as an ionization chamber. These embodiments (not shown) look similar to FIGS. 14-19, just thinner due to the more severe average camera-lens angle of ˜60°-70° as compared to the ˜45° angle in FIGS. 14-19.

For the above disclosed neutron beam monitors, several high neutron cross-section isotope loaded scintillators are available, such as Eljen EJ-254 or Saint-Gobain BC-454 which are both B10 loaded plastic PVT-based scintillators, or cerium activated Li6 doped silicate glass scintillators from Saint-Gobain, although Li6 doped plastics have also been fabricated. For the various neutron capture therapy (“NCT”) applications, including both BNCT and GdNCT (gadolinium-NCT), the neutron beams employed span the energy range from thermal-NCT to fast-NCT (also called FNT), but most NCT programs appear to be based on epithermal-NCT. Unfortunately all of these neutrons are also more damaging to the scintillator material than protons and/or photons, and therefore scintillator replacement would need to occur much more frequently. For this reason the internal calibration scheme employed in the above embodiments is important for the successful implementation of neutron beam monitors, and the fact that scintillator replacement and internal calibration could be accomplished within minutes would be even more beneficial for NCT than for proton or photon EBRT due to more frequent replacement. One method to prolong the useful lifetime of the boron doped scintillator, and therefore not have to replace it as often, is to integrate a motorized X-Y translation stage into the beam monitor enclosure structure and thereby translate the entire system in the X-Y plane in relatively small steps as required, thus moving it around the isocenter and lengthening the period between scintillator replacement—this strategy is conceptually similar to moving the scintillator-frame in small steps in the 6-way-cross via the previously described push-pull linear positioners.

A general complication associated with scintillators for neutron detection is that most neutron sources also generate gammas, and scintillators that detect neutrons will therefore also detect gammas. Most applications, be they medical imaging or homeland security, require neutron detection systems that can effectively discriminate between gammas and neutrons. The disclosed beam monitor embodiments in FIGS. 18 and 19 can effectively provide such discrimination for NCT applications such as BNCT and GdNCT, as well as for other applications such as homeland security. The method by which this can be achieved is to use two different scintillators, as configured in FIGS. 18 and 19, where for example the scintillator on one side (e.g., entrance window) such as 1862 and 1962 respectively might consist of either a thin BoPEN film or a conventional polyvinyl toluene (“PVT”) or polystyrene (“PS”) based gamma/ion scintillator, with the scintillator on the opposite side (e.g., exit window) such as 1860 and 1960 respectively being a neutron sensitive scintillator such as the boron loaded EJ-254 based PVT (˜5% natural boron) or BC-454 based PVT (˜5% natural boron, although 10% natural boron is also available). The method to separate the neutron generated image/signal from that produced by gammas is to digitally subtract the image/signal generated by the 1862 or 1962 scintillator from that generated by the 1860 or 1960 scintillator. Such a design will mimic or behave as though it has a high level of gamma to neutron discrimination.

In looking into scintillator damage by neutrons, the issue of radiation damage to the beam monitor cameras was also investigated. Experiments indicate that the slow radiation damage over a period of years to the patient viewing cameras in proton therapy treatment rooms is primarily due to neutrons. The main source of these neutrons is not from the proton beam system, although some neutrons are generated in the collimator, but from the patient's interaction and absorption of the proton beam itself—i.e. primarily where the proton beam stops at the tumor site inside the patient. Radiation damage to digital cameras has been studied extensively for the imaging sensors used in space astronomy (mostly CCDs), as well as for other situations in which high neutron fluxes are created and monitored by cameras such as for fusion research. It has been found that although shielding of cameras can be helpful, it is also not so straightforward. One solution is the direct cooling of the camera sensor to about −20° C. or colder, which also eliminates the camera as a heat source and thereby reduces thermal heating causing calibration drift not only of the camera, but also of the UV-LEDs, UV-photodiodes, and possibly even the scintillator response itself.

Camera sensors/electronics are prone to neutron damage because silicon is typically doped with boron to achieve p-type silicon. However, p-type silicon can also be produced by doping with gallium (“Ga”) instead of boron, and in this way fabricate radiation-hardened silicon devices. Both radiation-hardened and radiation-tolerant semiconductors, including CMOS image sensors and cameras are available from several sources, as such sensors and cameras are required for a number of applications including military, aerospace, scientific, and nuclear energy. With conventional boron doped silicon devices, the primary camera visual damage due to neutrons is the creation mostly of “bright” pixels in the silicon image sensor. The “bright” pixels caused by rad-damage are high dark-current pixels or “hot-pixels”. Some embodiments replace the cameras in the beam delivery room every couple of years. The majority of neutrons created are scattered in the proton beam momentum direction, which is towards the opposite side from where the patient is being irradiated and thus towards the back of the room. However the entire room is effected by the scattered neutron field and some neutrons will backscatter towards the cameras located at the beam nozzle exit and in front of the patient. Frequent internal calibration of the beam monitoring system will identify the radiation damaged pixels, and so their contribution to the image analysis can be conveniently eliminated by software. Partial neutron shielding of the cameras can be achieved by several means, including the use of boron doped transparent plastics in front of the camera body and lens, similar to commercially available 5% boron doped PVT plastic scintillators but without the addition of a fluor dopant. Since the cameras themselves are located out of the direct beam path, the entire light-tight camera box enclosure, excluding the entrance and exit window areas, can be fabricated out of a neutron shielding metal sheet such as a boron-aluminum alloy like BorAluminum from Ceradyne (˜4.5% to 8% by weight of B-10 isotope) or AluBor (10% by weight of natural boron) from S-DH, or a boron clad aluminum such as BORAL or BORTEC. Also boron composite plates made with boron fiber can be used. Alternatively, a number of small shielding plates can be strategically placed around each camera body. Another solution for shielding the front of the camera from neutrons is to use a thick, high boron content transparent borosilicate glass (e.g., 3-5% boron) in front of the camera lens, and maybe in front of the entire camera body. There are many borosilicate optical glasses, but Schott N-ZK7 (15% B2O3 by wt.), N-BK10 (13% B2O3 by wt.) and N-BK7 (10% B2O3 by wt., also referenced as Borkron) with 4.7%, 4.0% and 3.1% boron respectively (by weight), or Schott BOROFLOAT-33 with 4.0% boron (i.e., 13% B2O3) are all readily available as is Corning 7740 glass (Pyrex) which is 12.6% B2O3. However BOROFLOAT-33 being much more economical than other borated glasses is sold for neutron shielding in thicknesses up to ˜25 mm. It is noted that extremely high B2O3 and Gd2O3 glasses have been described in the patent literature such as Application PCT/JP2013/069578 which potentially would be more effective. Also a source of heavily doped boron and lithium polyethylene sheets, bricks and rods/cylinders is Shieldwerx (a division of Bladewerx LLC), which sells a 30% natural boron doped polyethylene product called SWX-210 (i.e., contains 1.87×1022 boron atoms per cm3) as well as a 7.5% natural lithium doped polyethylene product SWX-215. The disadvantage of boron doped neutron shielding materials is that each neutron captured by boron generates a 0.42 MeV gamma ray; however, lithium doped shielding materials do not produce any neutron capture gammas. However the lower neutron capture cross-section of Li6 compared to B10, means that a greater thickness of lithium doped material is required than similarly doped borated material.

In terms of neutron damage, at what point the cameras, or possibly just the silicon image sensors, would have to be replaced needs to be experimentally determined, but most likely it will be in years for proton or photon therapy since the internal calibration system can adjust for bad pixels in real-time as they occur. It is also noted that the larger the image sensor pixel size, the less prone it is to radiation damage. For the beamline monitors such as the 6-way-crosses in FIGS. 9-13, the camera can be moved a significant distance away from the beamline and hence the radiation field, with the detrimental effect on photon collection and spatial resolution minimized, by extending the optical system length. This can be achieved via the introduction of a relay lens system including a relay train assembly. The specific design will depend on the distance desired for extending the optical tube length of the camera system. Relay lenses are made to extend the viewing distance for remote viewing and operate by producing intermediate planes of focus. Collecting and dispensing optical images is done with focusing lenses which transport the light pattern via a relay lens or train of relay lenses. Some examples include periscopes, endoscopes, remote inspection and surveillance. A wide selection of relay lenses are commercially available.

Embodiments are directed to external beam radiation therapy (“EBRT”) related applications for both particle and photon radiation. For both types of EBRT, the embodiments are directed towards beam monitoring systems designed for use in either of two locations: (1) internal beam monitors located within the accelerator beam delivery system and therefore prior to the beam exiting the system nozzle or snout or collimator, or (2) external beam monitors located outside the accelerator beam delivery system after exiting the system nozzle or snout or collimator and thus positioned after the delivery system exit and in front of the patient.

Embodiments can further be used for a variety of industrial and scientific beam monitoring applications such as ion implantation accelerators (e.g., depending on ion, typically >0.3 MeV), and nuclear physics particle accelerators. Typically ion beam implantation will have the most stringent detector/monitor design requirements with regard to beam transparency, as the ion particle energies are frequently below 1 MeV and the particles themselves are typically highly ionized, heavy nuclei. Many accelerators used for nuclear physics also operate at relatively low to medium ion energies, so the same beam monitor concept in accordance to embodiments can be used for both applications. Some additional advantages of the described embodiments include the relative low cost of the beam monitor critical hardware, and the low cost lifetime operational/maintenance expense which includes the minimal overhead expense associated with the ultra-fast internal calibration system, as compared with the time consuming calibration cost for conventional systems. This benefit is also important for scientific applications (e.g. nuclear physics) that subject other detectors/monitors to costly maintenance and radiation damage replacement expenses.

The therapeutic benefits of embodiments of UFT beam monitors disclosed herein are particularly useful with “FLASH” irradiation therapy in which short pulses (˜0.5 second) of radiation are delivered at ultrahigh dose rates of ≥10 Gy/s (i.e., FLASH) compared to conventional dose rates of ≤0.03 Gy/s in single doses over a period of ≥60 seconds. FLASH radiotherapy may well result in a paradigm shift in the treatment of cancer as ultrahigh dose rates appear to increase the differential response between normal and tumor tissue, thus increasing the lethality to malignant cells while not significantly increasing damage to healthy cells. In order to monitor the FLASH beam in real-time, the much faster beam profile imaging time and readout capability, greatly improved intrinsic 2D position resolution, and the much higher dose rate capability of the described UFT beam monitors yields order-of-magnitude advantages when compared to conventional ionization chambers, and in this sense appears to be an unexpected enabling technology.

Embodiments include a transmissive ionizing-radiation beam monitoring system comprising an enclosure structure with an ultra-thin, dark colored or black exit window to an incident ionizing-radiation beam; at least one thin scintillator within the enclosure structure that is directly in an incident ionizing-radiation beam path and transmissive to the incident radiation beam; at least one ultraviolet (UV) illumination source within the enclosure structure facing the scintillator for internal system calibration; at least one UV photosensor within the enclosure structure positioned to monitor and calibrate the UV illumination source; and at least one machine vision camera within the enclosure structure located out of an incident ionizing-radiation beam path and comprising a camera body and lens having a projection of its optical axis oriented at an angle of incidence of 50±30 degrees to a surface of the scintillator.

Embodiments further include a computer system comprising a frame grabber to process and analyze image data streaming in real-time from the machine vision camera; a wired cable or wireless data interface connection between the machine vision camera and the computer system; a plurality of machine vision cameras surrounding a scintillator area, each of the plurality of cameras having a corresponding lens focused on and its field of view centered on one particular section of the scintillator area; a UV bandpass filter optically coupled in close proximity to each UV illumination source, the UV bandpass filter having maximum spectral transmission in a spectral region of maximum emission from the UV illumination source; wherein the UV illumination source comprises a UV Light Emitting Diode (LED) comprising an emission peak that corresponds to a strong absorption region of the scintillator with minimal luminous output in a scintillator emission region; wherein the computer system further comprises tracking a beam position and movement, and calculating a beam shape, a beam intensity profile, a beam fluence and external dosimetry in real-time from a streaming digital output of the machine vision camera; wherein the thin scintillator comprises biaxially-oriented polyethylene naphthalate (Bo PEN).

Embodiments include a transmissive ionizing-radiation beam monitoring system comprising a light-tight enclosure comprising an ultra-thin entrance window and an ultra-thin exit window to an incident ionizing-radiation beam, wherein the entrance window and the exit window are highly transmissive to the incident ionizing-radiation beam and comprise one of a dark colored or black polymer film or metal foil, or a composite polymer-foil combination; at least one thin scintillator within the light-tight enclosure that is directly in an incident ionizing-radiation beam path and transmissive to the incident radiation beam; at least one ultraviolet (UV) illumination source within the light-tight enclosure facing the scintillator for internal system calibration; at least one UV photosensor within the light-tight enclosure positioned to monitor and calibrate the UV illumination source; at least one machine vision camera within the light-tight enclosure comprising a camera body and lens system located out of an incident ionizing-radiation beam path; and a mirror in close proximity to each lens and located out of an incident ionizing radiation beam path and obliquely facing both the lens and the scintillator at an angle, wherein the machine vision camera and its associated close proximity mirror comprises a folded optical system configuration with respect to its view of the scintillator surface to reduce a thickness or depth of the light-tight enclosure.

Embodiments further include an ionizing-radiation beam optimized for treating cancer by external beam radiation therapy and comprise one of a particle beam of electrons, protons, ions or neutrons, or a photon beam of X-rays or gamma-rays; wherein the thin scintillator comprises a film or sheet of biaxially-oriented polyethylene naphthalate (BoPEN) falling within a thickness range between 1 μm and 300 μm; the scintillator comprising a roll-to-roll scintillator feed configuration wherein a scintillator film is wrapped around and stored on a small diameter feeder-spool located inside the enclosure and pulled across a beam axis transit window area onto a take-up spool that can be advanced by a stepper-motor rotating a take-up spool spindle to move a new section of scintillator film across the beam window area to replace a previously radiation damaged area; wherein one or both of entrance and exit windows are individually physically coupled by a thin frame to the thin scintillator such that a window/scintillator framed structure comprises a single replaceable module unit that can be accessed from outside the light-tight enclosure; further comprising multiple machine vision cameras surrounding a scintillator area, with each camera having the lens of its folded optical system focused and its field of view centered on one particular section of the scintillator area; wherein a computer system comprising a frame grabber combines data from the multiple cameras to track the beam position and movement across an entire scintillator area and computes the beam shape, a beam intensity profile, a beam fluence and external dosimetry in real-time from a streaming digital output of the machine vision cameras; further comprising multiple machine vision cameras surrounding a scintillator area, with each camera being in close proximity and optically coupled to a small mirror and thereby constituting a multi-camera folded optical system configuration so as to minimize the thickness of the light-tight enclosure; wherein some of the machine vision cameras operate at short exposure times while other machine vision cameras operate at long exposure times; wherein for a dual entrance-exit window/scintillator module system some of the machine vision camera proximity mirror folded optical system units are focused on an entrance scintillator area, with a remainder focused on an exit scintillator area, so that the computer system can calculate a beam angular divergence between the two scintillators while also improving on an accuracy and resolution of the beam position, movement, intensity profile, fluence and external dosimetry.

Embodiments include a transmissive ionizing-radiation beam monitoring system including a vacuum chamber structure with vacuum compatible flanges through which an incident ionizing-radiation beam enters and exits the monitoring system; at least one thin scintillator within the vacuum chamber structure that is transmissive to the incident ionizing-radiation beam and oriented at an angle greater than 10 degrees to a normal of the incident radiation beam; a machine vision camera within an arm or nipple or small enclosure at atmospheric/ambient pressure attached to the vacuum chamber structure by a flange attached to a vacuum-tight viewport window with a camera optical axis oriented at an angle of less than 80 degrees with respect to a normal of the scintillator; at least one UV illumination source facing the scintillator for internal system calibration; and at least one UV photosensor positioned to monitor and calibrate the UV illumination source.

Embodiments further include a multi-arm cross with the camera located within a nipple or small enclosure at ambient pressure that is attached to a flange of a viewport window on one arm of the cross; wherein the thin scintillator is mounted in a frame and attached to a shaft of a push-pull linear positioner that can be pushed or pulled or nudged through a beam area from one side or arm of the cross into an opposite side or arm; wherein the scintillator in its frame is oriented at an angle of 45±25 degrees to the normal of the incident radiation beam and the camera optical axis is oriented at an angle of 45±25 degrees to the normal of the scintillator; wherein the vacuum chamber structure is a 6-way cross with the camera located within a small enclosure attached to the flange of the viewport window on one arm of the cross, and a photomultiplier tube (PMT) or solid state photomultiplier (SSPM) such as a silicon photomultiplier (SiPM) located within a second small enclosure also at atmospheric/ambient pressure and attached to the flange of a second viewport window on the arm opposite the camera; wherein a first condensing lens is located in close proximity to the viewport window in the small enclosure at atmospheric/ambient pressure containing the PMT or SSPM, and a second condensing lens is located on the other side of the same viewport window in the vacuum chamber structure just below the scintillator frame, with the two condensing lenses separated by the viewport window but facing each other belly-to-belly to capture a relatively large solid angle of light from the scintillator and projecting it onto a light sensitive area of the PMT or SSPM; wherein a first scintillator in its frame is located in one arm of the 6-way cross and a second scintillator in its frame is located in the opposite arm of the 6-way cross, with each scintillator attached to its own push-pull linear positioner, and wherein the first and second scintillators do not have to be identical either in composition or in thickness; wherein the scintillator comprises a roll-to-roll scintillator feed configuration wherein a scintillator film is wrapped around and stored on a small diameter feeder-spool located inside the vacuum chamber structure and pulled across a beam axis transit area onto a take-up spool that can be advanced by a stepper-motor rotating the take-up spool spindle to move a new section of scintillator film across the beam axis transit area to replace a previously radiation damaged area.

Embodiments include a method of monitoring a beam of ionizing-radiation, including tracking of beam position, movement, intensity profile, beam fluence and external dosimetry, with a rapid internal calibration system, the method comprising: receiving the ionizing-radiation beam in a transmissive thin scintillator enclosed in a light-tight structure with entrance and exit highly transmissive to the incident radiation beam, a folded optical system comprising at least one mirror and one machine vision camera, at least one UV illumination source facing the thin scintillator and at least one UV photosensor positioned to monitor the UV source, a computer system comprising a frame grabber and a wired cable or wireless data interface between each machine vision camera and the computer system; creating a multitude of emitting photons some of which are captured by the machine vision camera folded optical system in real-time; causing a series of streaming images from each machine vision camera to the computer system; wherein the computer system processes and analyzes in real-time the image data streaming from the machine vision cameras to track the beam position, movement, intensity profile, beam fluence and external dosimetry; wherein the rapid internal calibration system periodically activates the UV-source to illuminate the thin scintillator over its entire active area while the UV-photosensor monitors the UV-source to correct for any UV-source signal drift or instability over time, while the machine vision camera folded optical system streams images of the scintillator photon emission over its active area; wherein the computer system compares the machine vision camera digital image output to previously stored digital image output taken of the thin scintillator under the same conditions and in this manner monitors the system stability over time and internally calibrates the system for small changes in performance, and signals when a change in the scintillator component is required according to pre-programmed guidelines, or flag if any other hardware or software problems are detected.

Several embodiments are specifically illustrated and/or described herein. However, it will be appreciated that modifications and variations of the disclosed embodiments are covered by the above teachings and within the purview of the appended claims without departing from the spirit and intended scope of the invention.

Claims

1. A transmissive ionizing-radiation beam monitoring system comprising:

an enclosure structure with an ultra-thin window to an incident ionizing-radiation beam;
at least one thin scintillator within the enclosure structure that is substantially directly in an incident ionizing-radiation beam path;
at least one ultraviolet (UV) illumination source within the enclosure structure;
at least one UV photosensor within the enclosure structure; and
at least one machine vision camera within the enclosure structure.

2. The transmissive ionizing-radiation beam monitoring system of claim 1, further comprising:

a computer system that processes and analyzes image data streaming in real-time from the machine vision camera.

3. The transmissive ionizing-radiation beam monitoring system of claim 1, further comprising:

a UV bandpass filter optically coupled in close proximity to each UV illumination source, the UV bandpass filter having maximum spectral transmission in a spectral region of maximum emission from the UV illumination source.
Patent History
Publication number: 20200108278
Type: Application
Filed: Nov 27, 2019
Publication Date: Apr 9, 2020
Applicant: Integrated Sensors, LLC (Ottawa Hills, OH)
Inventor: Peter S. FRIEDMAN (Ottawa Hills, OH)
Application Number: 16/697,439
Classifications
International Classification: A61N 5/10 (20060101); A61N 5/00 (20060101);