Objective. This work presents a method for enhanced detection, imaging, and measurement of the thermal neutron flux. Approach . Measurements were performed in a water tank, while the detector is positioned out-of-field of a 20 MeV ultra-high pulse dose rate electron beam. A semiconductor pixel detector Timepix3 with a silicon sensor partially covered by a ⁶ LiF neutron converter was used to measure the flux, spatial, and time characteristics of the neutron field. To provide absolute measurements of thermal neutron flux, the detection efficiency calibration of the detectors was performed in a reference thermal neutron field. Neutron signals are recognized and discriminated against other particles such as gamma rays and x-rays. This is achieved by the resolving power of the pixel detector using machine learning algorithms and high-resolution pattern recognition analysis of the high-energy tracks created by thermal neutron interactions in the converter. Main results . The resulting thermal neutrons equivalent dose was obtained using conversion factor (2.13(10) pSv·cm ² ) from thermal neutron fluence to thermal neutron equivalent dose obtained by Monte Carlo simulations. The calibrated detectors were used to characterize scattered radiation created by electron beams. The results at 12.0 cm depth in the beam axis inside of the water for a delivered dose per pulse of 1.85 Gy (pulse length of 2.4 μ s) at the reference depth, showed a contribution of flux of 4.07(8) × 10 ³ particles·cm ⁻² ·s ⁻¹ and equivalent dose of 1.73(3) nSv per pulse, which is lower by ∼9 orders of magnitude than the delivered dose. Significance . The presented methodology for in-water measurements and identification of characteristic thermal neutrons tracks serves for the selective quantification of equivalent dose made by thermal neutrons in out-of-field particle therapy.

This work investigates the operational acquisition time limits of Timepix3 and Timepix2 detectors operated in frame mode for high-count rate of high deposited energy transfer particles. Measurements were performed using alpha particles from a ²⁴¹Am laboratory source and proton and carbon ion beams from a synchrotron accelerator. The particle count rate upper limit is determined by overlapping per-pixel particle signals, identifiable by the hits per pixel counter > 2, indicating the need to decrease acquisition time. On the other hand, the lower limit is the time required to collect the particle deposited charge while maintaining spectral properties. Different acquisition times were evaluated for an AdvaPIX Timepix3 detector (500 μm Silicon sensor) with standard per-pixel Digital to Analog Converter (DAC) settings and a Minipix Timepix2 detector (300 μm Silicon sensor) with standard and customized settings the pulse shaping parameter and threshold. For AdvaPIX Timepix3, spectra remained accurate down to 100 μs frame acquisition time; at 10 μs, loss of collected charge occurred, suggesting either avoiding this acquisition time or applying a correction. Timepix2 allowed acquisition times down to 100 ns for single particle track measurements even for high energy loss, enabled by a new Timepix2 feature delaying shutter closure until full particle charge collection. This work represents the first measurement utilizing Timepix-chips pixel detectors in an accelerator beam of clinical energy and intensity without the need to decrease the beam current. This is made possible by exploiting the short shutter feature in Timepix2 and a customized per-pixel energy calibration of the Timepix2 detector with a larger discharging signal value which allowed for shorter Time-over-Threshold (ToT) signal. These customized settings extend the operation of the pixel detectors to higher event rates up to 10⁹ particles/cm²/s.

Objective. Given the increased interest in incorporating linear energy transfer (LET) as an optimization parameter in intensity-modulated proton therapy (IMPT), a solution for experimental validation of simulations and patient-specific quality assurance (PSQA) in terms of proton LET is needed. Here, we present the methodology and results of LET spectra measurements for spread-out Bragg peak (SOBP) and IMPT plans using a miniaturized pixel detector Timepix3. Approach. We used a MiniPIX Timepix3 detector that provides single-particle tracking, type-resolving power, and spectral information while allowing measurement in quasi-continuous mode. We performed measurements for SOBP and IMPT plans in homogeneous RW3 and heterogeneous CIRS head phantoms with reduced beam current. An artificial intelligence-based model was applied for proton identification and a GPU-accelerated FRED Monte Carlo (MC) code was applied for corresponding MC simulations. Main results. We compared the deposited energy and LET spectra obtained in mixed radiation fields from measurements and MC simulations. The peak positions of deposited energy and LET spectra for the SOBP and IMPT plans agree within the error bars. Discrepancies exceeding the error bars are only visible in the logarithmic scale in high-energy deposition and high-LET tails of the distributions. The mean relative difference of dose-averaged LET values between measurements and MC simulations for individual energy layers is about 5.1%. Significance. This study presents a methodology for assessing radiation quality in proton therapy through energy deposition and LET spectra measurements in uniform and clinical IMPT fields. Findings show an agreement between experimental data and MC simulations, validating our approach. The presented results demonstrate the feasibility of a commercially available Timepix3 detector to validate LET computations in IMPT fields and perform PSQA in terms of LET. This will support the implementation of LET in treatment planning, which will ultimately increase the effectiveness of the treatment.

With the increasing use of proton therapy, there is a growing emphasis on including radiation quality, often quantified by linear energy transfer, as a treatment plan optimization factor. The Timepix detectors offer energy-sensitive particle tracking useful for the characterization of proton linear energy transfer. To improve the detector's performance in mixed radiation fields produced in proton therapy, we customized the detector settings and performed the per-pixel energy calibration. The detection threshold and per-pixel signal shaping time (I_Krum current) were customized, and energy calibration was performed for MiniPIX Timepix3. The detector calibration was verified using α source and clinical proton beams, as well as Monte Carlo simulations. The effects on the detector's performance, in terms of spectral saturation and pixel occupancy, were evaluated. Measurements with proton beams showed a good agreement with simulations. With the customized settings, the measurable energy range in the detector data-driven mode was extended, and the signal duration time was reduced by 80%, while the yield of pixel time occupancy reduction depends on the number of occupied pixels. For performed measurements with proton beams, the number of occupied pixels was further reduced up to 40% due to the increased threshold. Customized detector configuration of the Timepix3 detector allowed for reduced pixel occupancy and mitigation of signal saturation in a data-driven mode without significantly interfering with the energy deposition measurement. The presented approach enables the extension of the operational range, including higher intensities and mixed-radiation fields in particle radiotherapy environments.

Objective. This study aims to assess the composition of scattered particles generated in proton therapy for tumors situated proximal to some titanium (Ti) dental implants. The investigation involves decomposing the mixed field and recording Linear Energy Transfer (LET) spectra to quantify the influence of metallic dental inserts located behind the tumor. Approach. A therapeutic conformal proton beam was used to deliver the treatment plan to an anthropomorphic head phantom with two types of implants inserted in the target volume (made of Ti and plastic, respectively). The scattered radiation resulted during the irradiation was detected by a hybrid semiconductor pixel detector MiniPIX Timepix3 that was placed distal to the Spread-out Bragg peak. Visualization and field decomposition of stray radiation were generated using algorithms trained in particle recognition based on artificial intelligence neural networks (AI NN). Spectral sensitive aspects of the scattered radiation were collected using two angular positions of the detector relative to the beam direction: 0° and 60°. Results. Using AI NN, 3 classes of particles were identified: protons, electrons & photons, and ions & fast neutrons. Placing a Ti implant in the beam’s path resulted in predominantly electrons and photons, contributing 52.2% of the total number of detected particles, whereas for plastic implants, the contribution was 65.4%. Scattered protons comprised 45.5% and 31.9% with and without metal inserts, respectively. The LET spectra were derived for each group of particles identified, with values ranging from 0.01 to 7.5 keV μm⁻¹ for Ti implants/plastic implants. The low-LET component was primarily composed of electrons and photons, while the high-LET component corresponded to protons and ions. Significance. This method, complemented by directional maps, holds the potential for evaluating and validating treatment plans involving stray radiation near organs at risk, offering precise discrimination of the mixed field, and enhancing in this way the LET calculation.