TABLE 1.
Electron FLASH dosimetry with modified LINACs.
| Detector class | Detector model(s) | Reference | Uses a | Calibration | Compared against? | Agreement? (Y/N) | Modified LINAC | Energy (MeV) | DR/repetition rate | DPP | Source‐detector distance | FS and applicator | Comments |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Diamond | microDiamond (PTW, SN122323, and SN122325) read with Unidos webline electrometer (PTW) b | Dal Bello et al. 23 | TAD, BC (profiles, output vs. gantry angle, and PDDs) | Cross‐calibrated against EBT3 films | EBT3 Gafchromic Film (calibrated 0.5‐20 Gy in 16 MeV CONV), HD‐V2 films (Ashland, calibrated 1–100 Gy in CONV), IC profiler (calibrated in 16 MeV CONV), and corrected Advanced Markus IC. Mainly comparing between CONV and FLASH characteristics however | Y | Varian Truebeam SN 1001 | 16 | Average DR at isocenter = 256 Gy/s, tested upstream at SSD = 59 cm for DR = 837 Gy/s // Tested 20–200 Hz | Max DPP at isocenter = 1.28 Gy/pulse | Isocenter (in 3 cm depth in RW3 phantom) | Mostly 10 × 20 cm2, tested 4 × 4–25 × 25 cm2 | Only SN 122325 was found to be linear up to 1.5 Gy/pulse; the other was not. Films were used to verify PDDs/Profiles/Output, IC Profiler was only for profiles, and Advanced Markus chamber was only for output factors. |
| Diode | EDGE Diode SNC 1118 read with modified PC electrometer from SNC 1014 | Rahman et al. 24 | TAD, BC (PDDs), DPP | Cross calibrated against film dose measured at 2 cm depth in solid water | TAD—EBT‐XD Gafchromic Film and Ion Chamber in bremsstrahlung tail // BC—EBT‐XD Gafchromic Film // DPP—Exradin W1 scintillator | Y (within a specific ranges) | Varian Clinac 2100 C/D | 10 | 30–180 Gy/s // 360 Hz | 0.05–0.78 Gy/pulse | Isocenter | N/A | Modifications to electrometer performed by manufacturer. DPP was able to be measured, but the USB connection was not fast enough to transfer the data in real time |
| Sloop et al1 17 | DPP | N/A | N/A | N/A | Varian 2100EX and Trilogy | 10 | 432–72 Gy/s at isocenter // 360‐30 Hz | ∼ 1.2 Gy/pulse | Isocenter in 5 mm depth in solid water | Open | Compared two LNAC systems | ||
| EDD 2–3G Diode (IBA) counted with microcontroller b | Lempart et al., 25 Konradsson et al. 26 (same system) | PC | N/A | External monitor chamber connected to a digital oscilloscope (RBT2004 Rohde & Schwarz) (Konradsson) | Y | Elekta Precise | 10 | Tested from 30–300 Gy/s // 200 Hz | Tested from 0.18–1.9 Gy/pulse | Isocenter (off axis), also 19, 37, 53 cm | 10 × 10 cm2 applicator and 40 × 40 cm2 open | At the edge of the field. MCU measures the rising edge of the signal. Different MCUs were used (Arduino Due and Atmega 238) | |
| EDP 20–3G Diode (IBA) b | Lempart et al. 25 | TAD | N/A | EBT3 Gafchromic film | Linear with DPP and dose | Elekta Precise | 10 | Tested from 30–300 Gy/s // 200 Hz | Tested from 0.18–1.9 Gy/pulse | 19, 37, and 53 cm | 40 × 40 cm2 | Results show this detector could be calibrated to measure dose in FLASH | |
| TLD/OSLD | 3 Types of TLDs: LiF:Mg,Ti, LiF:Mg,Cu,P, CaF2:Tm // 2 Types of OSLDs: Al203:C, BeO all from Luxel in 3D PLA holder | Motta et al. 27 | TAD | Read background signal, irradiated with a built‐in known source on the reader for a set time, read again. Repeat for several times to make a calibration curve (one for low dose and one for high dose). Relate to dose in CONV electron beams. | N/A | N/A | Varian Truebeam | 16 | DR: 200–3 × 105 Gy/s // 200 Hz | 8.3 × 10−4–1.255 Gy/pulse (isocenter) | Isocenter (in 3 cm depth in RW3 phantom) | 10 × 10 cm2 | N/A |
| Round TLD‐100 LiF:Mg:Ti chips from Thermo Fisher | Jorge et al. 28 | TAD | Calibrated in terms of dose to Co‐60 at institute of radiation physics. Traceable to METAS (primary standard) | EBT3 Gafchromic Film and Alanine (calibrated in CONV) | Y, closest agreement to film, farther from expected than Alanine | Varian Trilogy | 16.6 | Average of 150 Gy/s // 90 Hz | Average of 1.5 Gy/pulse | 19.2 cm | 4 × 4 cm2 | Applied corrections at time of reading with uncertainty of 2.9% | |
| Radio—luminescence | Exradin W1 Scintillator b | Ashraf et al. 29 | TAD, DPP | Measured 2 calibration factors: gain and Cherenkov to light ratio (Frontbonne et al. 2002) | TAD—Gafchromic film | Y | Varian C‐Series | 10 | 50–350 Gy/s // 60–360 Hz | 0.3 ‐ 1.1 Gy/pulse | Range tested: 77–130 cm | Small circular FS 1 cm–4 cm and calibration in 40 × 40 cm2 open field | Radiation hardness issues |
| Exradin W2 Scintillator (1 × 1 mm and 1 × 3 mm) with modified MAX SD signal processor b | Oh et al. 30 | TAD | MAX SD finds a Cherenkov to light ratio from the green and blue color channels | EBT‐XD Gafchromic film and Advanced Markus Chamber plane parallel | Y, within 2%. Markus agreed with film within 5% up to 0.4 Gy/pulse | Varian Clinac 23EX | 16 | 180 Gy/s at isocenter // 18–180 Hz | Max 3.6 Gy/pulse (for one pulse) | Range tested: 61.6–132 cm | 10 × 10 cm2 no applicator | Vendor of MAX SD modified the device for them. The 1 × 3 mm was only linear up to 0.4 Gy/pulse (multiple pulses) or 0.8 Gy/pulse (one pulse). 1 × 1 mm was linear with dose to the max of the system. Optical fiber is damaged 2% per kGy | |
| DoseOptics iCMOS Camera | Rahman et al. 31 | TAD, BC (surface profiles and PDDs), RPP | Found conversion factor from counts to dose with EBT‐XD film measurements. Corrected for camera angle with MATLAB transformation | EBT‐XD Gafchromic film | Y—TAD—within 3% after ramp up // BC—Scintillation agreed better by 10% in gamma analysis (3%/3 mm) | Varian Clinac 2100 C/D | 10 | 50–400 Gy/s // 60–360 Hz | 0.18 ‐ 0.91 Gy/pulse | Camera out of field (tested two configurations) | 40 × 40 cm2 | Camera was time gated to the KlyV signal from the LINAC. Captured Cherenkov emission from a solid water phantom and scintillation from a rare earth screen on top of a phantom. Scintillation was only confirmed to work properly in all aspects. Could resolve individual pulses. | |
| Ashraf et al. 32 | BC (3D dose distribution), RPP | N/A | EBT‐XD Gafchromic film | Y—Reconstructed Profiles (besides one) agreed 100% with gamma 2%/2 mm. Reconstructed PDD agreement besides buildup | N/A | N/A | 50–70 Gy/s // 60 Hz | N/A | Camera out of field | 40 × 40 cm2 and 1.5 cm circular field | Used bandpass filter to ignore Cherenkov. Could reconstruct individual pulses (time‐gated). Imaged water tank with quinine sulfate solution. Processing steps required to reconstruct median filter, flatfield, alignment, symmetrization, and Abel inversion | ||
| CMOS Camera—Quad Channel CXP‐12 GigaSens (Concurrent EDA) (with and without intensifier) | Rahman et al. 33 | BC (surface profiles), RPP | Dark field image subtraction, flat field correction, 3D median filtered. Did not calibrate to dose. | N/A | N/A | Varian Clinac 2100 C/D | 10 | Mean DR = 240 and 260 Gy/s for the 1 and 1.5 cm cutout //360 Hz | 0.66 Gy/pulse for 1 cm cutout, 0.72 Gy/pulse for 1.5 cm cutout | Camera out of field | 1 and 1.5 cm diameter cutouts | The camera was gated to the SYNC signal from the LINAC. SNR increased with intensifier, but SR decreased. Was able to resolve individual LINAC pulses but not in real time (required calibrations after). Cherenkov profile was larger than the actual field. | |
| HC‐120 Series PMT (Hamamatsu) coupled to optical fiber (Thorlabs) fed into a BNC cable, digitized with 3000 series Picoscope (PicoTechnology) interfaced with MATLAB b | Rahman et al. 31 | PC | N/A | N/A | N/A | Varian Clinac 2100 C/D | 10 | 50–300 Gy/s at the isocenter // 60–360 Hz | 0.81 Gy/pulse | Edge of field at isocenter distance | N/A | N/A | |
| Coincidence scattered radiation detector (DoseOptics) (Remote trigger unit RTU) b | Ashraf et al., 29 Rahman et al. 16 (same system) | PC | N/A | N/A | N/A | Varian Clinac 2100 C/D | 10 | 50–300 Gy/s at the isocenter // 60–360 Hz | Up to 1.1 Gy/pulse | Many distances | Many FS | Two BGO scintillators and two silicon photomultipliers | |
| Hyperscint RP100 Scintillator system b | Poirier et al. 20 | TAD | Software system prompts user for measurements to remove Cherenkov, background, and relate to dose from CONV irradiation | EBT‐XD Gafchromic Film and CC04 Pinpoint Ion Chamber IBA (relative) | Y within 4% of films | Varian 21EX | 16 | 100 Gy/s at 55 cm SSD, 34 Gy/s at isocenter // 18–180 Hz | Tested 0.55, 0.51, 0.27, 0.20 Gy/pulse | Tested 55, 60, 80, 100 cm SSD | N/A | Dose calibration must be done every time the detector is disconnected from the reader. Software has a PC feature as well, but the readout system causes errors | |
| Alanine | Alanine pellets read with Bruker EPR Spectrometer | Jorge et al. 28 | TAD | Calibrated in CONV on an Elekta Synergy against ion chamber traceable to METAS | TLD and EBT3 Gafchromic film | Y, alanine was closest to the expected dose. | Varian Trilogy | 16.6 | Average of 150 Gy/s // 90 Hz | Average of 1.5 Gy/pulse | 19.2 cm | 4 × 4 cm2 | Used a custom phantom to test film, TLD, and alanine together. Lacked absolute reference. |
| Ion chamber | IC Profilier Sun Nuclear b | No et al. 34 | BC (profiles) | Calibrated with 5 V on the grid to avoid saturation | EBT‐XD Gafchromic Film | Y | Varian Trilogy | 17.5 | 36.82–112.83 Gy/s // 180 Hz | 0.21 ‐ 0.63 Gy/pulse | 59–100 cm SSD | Tested 3, 10, and 20 cm circular custom‐made cutouts | N/A |
| Poirier et al. 21 | BC (profiles) | Calibrated with the lowest voltage possible to avoid damage | N/A | N/A | Varian 21EX | 16 | 2067 Gy/min at the monitor unit chamber | N/A | 165 cm SSD to reduce signal | N/A | N/A | ||
| Wellhofer Model CC01 b | Kyle et al. 35 | TAD | Calibrated in CONV 9MeV beam | Previously verified against film, scintillator, and OSLD (see comment) | Y | Varian Truebeam | 10 | 142 Gy/s and 59 Gy/s (depending on SSD) // 360 Hz | N/A | 62.5 and 100 cm SSD | 40 × 40 cm2 or 20 × 20 applicator | For verification of the IC, they reference an abstract presented at AAPM 2021 by Duzenli et al. | |
| Advanced Markus Plane Parallel (PTW 34045) b | Oh et al. 36 | TAD, BC (Profiles and output factors) | Found charge to dose curve and corrected recombination using models (see comment) | EBT3 Gafchromic Film and OSLDs (AL2O3:C) | Y, and the film/OSLDs were within 3% agreement | Varian Clinac 23EX | 16 | 500 Gy/s at 59 cm | 1.27 Gy/pulse at isocenter | Range tested (isocenter, 59 cm, etc.) | 10 × 10 cm2, no cone | Tested models for correcting for recombination (Boag with and without extension and logistic, which worked the best). Found linearity up to 0.1 Gy/pulse then saturation. | |
| CC04 Pinpoint IBA b | Poirier et al. 21 | Estimate TAD (see comment) | N/A | EBT‐XD Gafchromic Film | Y | Varian 21EX | 16 | 2067 Gy/min at the monitor unit chamber | N/A | SSD = 100 cm, 10 cm depth | 25 × 25 cm2 | Found a ratio of the signal in the bremsstrahlung tail from CONV to FLASH, applied this ratio to the dose in standard conditions in CONV. Measured a recombination factor of 1.06, outside of TG‐51 tolerance. | |
| Film |
Gafchromic EBT‐XD Ashland Gafchromic EBT3 Ashland |
Oh et al., 30 Cetnar et al., 37 No et al., 34 Petusseau et al., 38 Rahman et al., 31 Sloop et al., 17 Jorge et al., 28 Konradsson et al., 26 Garty et al., 39 Adrian et al., 40 Deut et al., 41 Konradsson et al., 42 Kyle et al., 35 Lempart et al., 25 Levy et al., 43 Szpala et al., 44 Xie et al. 45 | TAD, BC (PDDs and profiles) | Calibrated in CONV in a relavent dose range, usually following output check with ion chamber | Considered gold standard. Some compared to TLD, OSLD, Alanine, and ion chambers | N/A | Many different LINACs utilized | Range of energies | CONV–600 Gy/s observed | CONV–27 Gy/pulse observed | Range of setups | N/A | |
| OC‐1 OrthoChromic film for > 20 Gy | Garty et al. 39 | Varian Clinac 2100C | 9 | > 600 Gy/s // 180–360 Hz | >1–3 Gy/pulse | 20–30 cm SSD | 2 cm diameter beam projected at 20 cm SSD | ||||||
a
Uses of detectors are one of the following options: Total absorbed dose (TAD), beam characteristics (BC), dose per pulse (DPP), reading per pulse (RPP, same as DPP but not calibrated), or pulse counting (PC).
bSignifies detector systems that could measure online (i.e., in real‐time). Note that the detector may be in real‐time, but it does not have this symbol due to the readout system restricting its temporal resolution.