Table of contents

Micro-Mini & Nano Dosimetry and Innovative Technologies in Radiation Oncology - MMND ITRO 2020

The biannual Micro-Mini & Nano Dosimetry and Innovative Technologies in Radiation Oncology (MMND ITRO 2020) international conferences organized by the Centre for Medical Radiation Physics (CMRP), University of Wollongong jointly with Memorial Sloan Kettering Cancer Center (MSKCC), New York were hold on the 10th-16th February 2020 at the Novotel Wollongong Northbeach, Wollongong, NSW, Australia. During MMND ITRO 2020 we celebrated 20^th Anniversary of these biannual workshops.

Continuing the series of biennial meetings, the Mini- Micro- Nano- Dosimetry (MMND) conference brought together international and Australian radiation oncologists, medical physicists, radiation scientists and nanomedicine experts. Talks were focused on new achievements in radiobiology of radiation therapy including particle therapy and synchrotron MRT, diffused alpha emitters radiation therapy (DaRT) in combination with immunotherapy, radiomics and artificial intelligence as applied to radiation therapy and imaging, advanced dosimetry in radiotherapy, diagnostic imaging and diagnostic radiology for physicists. Special half-day sessions were led by experts from MSKCC on New Technologies in Brachytherapy by Prof. Michael Zelefsky MD, and Radiomics and Machine Learning Models by Prof. Joseph Deasy PhD.

Innovative Technologies in Radiation Oncology (ITRO) led by Dr Josh Yamada, MD was devoted to the clinical implementation of new technologies in X-ray therapy, brachytherapy and particle therapy. ITRO 2020 focused on integrating advanced imaging into radiation therapy including MRI guided radiotherapy, MRI for tissue radiation response prediction, imaging in prediction of patient response to immunotherapy, SBRT, SRS, hypofractionation and particle therapy as well as applications of radiomics and artificial intelligence in radiation oncology and radiology.

All papers published in this volume of Journal of Physics: Conference Series have been peer reviewed through processes administered by the Editors. Reviews were conducted by expert referees to the professional and scientific standards expected of a proceedings journal published by IOP Publishing.

• Type of peer review: Single-blind / Double-blind / Triple-blind / Open / Other (please describe)

Most papers included in Proceedings are based on Invited Talks. Open peer review of all papers was carried out by 18 experts in a filed in average 2 reviewers per each paper.

• Conference submission management system:

Conference submission site can be found at the conference site https://cmrp.events/at section "Participate" https://cmrp.events/abstract-submission/ all abstract and papers were submitted by email

• Number of submissions received:

Received: 40

• Number of submissions sent for review:

Sent to review :36

• Number of submissions accepted:

Accepted: 36

• Acceptance Rate (Number of Submissions Accepted / Number of Submissions Received X 100):

90%, most of the talks were presented by invited speakers

• Average number of reviews per paper:

2

• Total number of reviewers involved:

18

• Any additional info on review process:

After a final review all papers went through additional editing and formatting where needed

• Contact person for queries:

Prof Anatoly Rozenfeld anatoly@uow.edu.au, MMND ITRO General Chair

This paper introduces Varian portal dosimetry and IBA dolphin detector. Clinical machine QA examples were presented including DLG measurement and monthly beam constancy check. For patient-specific QA, both detectors generated similar passing rates to film and were able to identify an erroneous delivery with 3mm MLC offset.

Beam monitoring in particle therapy is a critical task that, because of the high flux and the time structure of the beam, can be challenging for the instrumentation. Recent developments in thin silicon detectors with moderate internal gain, optimized for timing applications (Ultra Fast Silicon Detectors, UFSD), offer a favourable technological option to conventional ionization chambers. Thanks to their fast collection time and good signal-to-noise ratio, properly segmented sensors allow discriminating and counting single protons up to the high fluxes of a therapeutic beam, while the excellent time resolution can be exploited for measuring the proton beam energy using time-of-flight techniques. We report here the results of the first tests performed with UFSD detector pads on a therapeutic beam. It is found that the signal of protons can be easily discriminated from the noise, and that the very good time resolution is confirmed. However, a careful design is necessary to limit large pile-up inefficiencies and early performance degradation due to radiation damage.

Treatment planning in hadron therapy uses Monte Carlo techniques that require total nuclear reaction cross sections for a variety of interacting systems and over a wide range of energies. Current empirical models are tuned to reproduce cross sections for stable isotopes, but in some cases, overestimate measured cross sections by 20%. Here we discuss an alternative approach: the optical limit of Glauber theory, which describes existing measurements better and offers more reliable predictions for the many systems where there is no experimental information. We also suggest how existing implementations of empirical models within GEANT4 could be improved. Ultimately, these calculations should contribute to improvements in planning of hadron therapy treatments.

University of Tsukuba is developing a new TPS for boron neutron capture therapy (BNCT) equipped with Monte Carlo dose-calculation engine based on Particle and Heavy Ion Transport code System PHITS. It is currently in the process of extending its adaptation to other radiotherapy beams. For this extension, not only physical doses but also their relative biological effectiveness (RBE) must be evaluated for various radiotherapy in the same framework. Frequent and dose probability densities of lineal energy, y, are the key quantities in the RBE estimation, and they must be precisely evaluated for various locations in a patient. In this study, the probability densities of y for a site diameter of 0.564 µm were calculated for X-ray, proton, carbon-ion, and BNCT beams with appropriate geometry settings using the microdosimetric function implemented in PHITS, and they were converted to the corresponding RBE-weighted doses using the microdosimetric kinetic model. The accuracy of the calculated data were well verified by several experimental data, indicating the adequacy of the use of PHITS and microdosimetric kinetic model in the dose-calculation engine for TPS applicable to various radiotherapy.

Clinical trials on carbon ion radiotherapy (C-ion RT) for the prostate cancer were conducted at the NIRS for over 20 years. The results so far have been quite satisfactory in both toxicity and tumour control. In addition, advancement of hypofractionation has been successfully progressed and 12-fraction C-ion RT could be established. To obtain the outcomes of multi-institutional clinical trials, a study group, J-CROS, was organized and conducts several multicenter clinical trials including the prospective study for the high-risk prostate cancer. It is expected that the outcomes of these clinical studies would result in widening the coverage of National Health Insurance for the C-ion RT. Technical progression has been also achieved at the NIRS. The scanning irradiation became available in 2011 and all the prostate cancers are treated with scanning at present. A new clinical trial of 4-fraction C-ion RT for the locally recurrent prostate cancer has been started using dose painting of scanning beam. The rotating gantry of heavy ion beam was also installed. These technical progressions enable to perform the intensity modulated ion therapy with carbon beam alone or with various ion species.

A microdosimetric characterization of the 62 MeV proton beam line of CATANA has been performed all along the Spread Out Bragg Peak with three different detectors. Two silicon detectors and a Tissue Equivalent Proportional Counter measured at approximately the same depths of the SOBP. The TEPC is a new miniaturized gas counter developed at the Legnaro National Laboratories of INFN, modified to work without gas flow. The first silicon detector has been developed at the Politecnico of Milano and it is a monolithic telescope composed by a matrix of 2 µm thick cylindrical diodes with a diameter 9 µm. that compose the ΔE layer. The E and ΔE layers are fabricated on a single substrate of silicon. The third detector is the MicroPlus probe developed at the CMRP - University of Wollongong, it is an array of 3D sensitive volumes each with dimension 30x30 µm and 10 µm thick fabricated on SOI. Measurements performed with the three detectors are presented and discussed.

In the present work, we reported on the use of a new 2D array of diodes, the Duo, for dosimetry of small beams produced with a CyberKnife system, and shaped with a novel multi-leaf collimator, the InCise 2.

Multiple vendors are now offering real-time MRI-guided radiotherapy systems. Quality assurance of the imaging and radiation isocentre alignment is an essential component of the QA programme for any linear accelerator used for delivering image-guided radiotherapy. In this work, the authors describe the design and feasibility testing of a device that combines a high resolution monolithic silicon strip detector with an MRI visible phantom for characterisation of optical, MR imaging and radiation isocentre for inline MR-guided radiotherapy systems.

In radiotherapy practice, 1D and 2D dosimeters are used for dose verification prior to patient treatment. Along with high accuracy and precision of dose measurements that these dosimeters provide, acquisition of dose deposition data in three dimensions requires extrapolation of measured data. Development of a 3D dosimeter would provide continuous information of dose distribution in matter. In this work, NaCl 3D crystal has shown that radiation deposition can be imaged using blue laser stimulation in two dimensions. It was further shown that the intensity of collected signal has near – linear dose dependence, however complete signal readout is required, to compensate for gradual signal collection at different depths along the profile of the stimulating laser beam, due to attenuation of the beam within the crystal. A method to extend dose measurement to three dimensions using imaging is proposed.

Image segmentation is a computer vision task aiming to establish a probabilistic mapping between individual pixels (2D) or voxels (3D) in an input image and a set of predefined semantic categories with reference to domain-specific knowledge. When applied to medical images, e.g. Magnetic Resonance Imaging (MRI), it allows delineation between healthy and abnormal tissue. Despite challenges due to lesion morphological heterogeneity, segmentation of brain tumours has the potential to streamline otherwise time-consuming manual annotation. Whereas brain tumour segmentation has continually advanced incorporating innovative deep learning methods, heuristics normally employed by radiologists have often been neglected. The focus of nearly all tumour segmentation articles thus far on 3D isotropic research-grade scans has also led to results of unknown generalisability to hospital-quality data. In order to address these gaps, this study has coalesced modern deep learning methods and clinical-driven priors into an optimised segmentation pipeline evaluated on clinical data at a large neurology and neurosurgery tertiary centre.

Photon excitation at selected energies following beta/gamma irradiation is found to significantly reduce the supralinearity of the thermoluminescence dose response of composite peak 5 in LiF:Mg,Ti (TLD-100). Following a dose of 100 Gy, photon excitation at an energy of 5 eV and fluence of 1.3 × 10¹⁸ ph cm⁻² reduces the normalized TL efficiency from 2.8 to 1.9 and at 400 Gy from 3.5 to 2.7. Excitation by photons of energy 3.65 eV (10¹⁹ ph cm⁻²) reduces the normalized efficiency of glow peak 5a (a low temperature satellite of peak 5) at 100 Gy from 2.9 to a value of 0.95 thereby resulting in a linear dose response. The high values of dose of 100 Gy and 400 Gy well beyond the normal dose range of clinical radiation therapy were chosen for demonstrative purposes in order to evaluate the likelihood of success of the proposed technique. Additional experiments are underway to determine the photon levels of fluence, which will result in a linear dose response for both glow peaks 5, and 5a in the range of doses 1-30 Gy of interest to radiotherapy and intraoperative electron therapy.

The rapidly increasing amount and complexity of data in healthcare, the pace of published research, drug development, biomarker discovery, and clinical trial enrolment in oncology renders AI an approach of choice in the development of machine assisted methods for data analysis and machine assisted decision making.

Machine learning algorithms, and artificial neural networks in particular, drive recent successes of AI in oncology. Performances of AI driven methods continue to improve with respect to both speed and precision thus leading to a great potential for AI to improve clinical practice. But the acceptance and a lasting breakthrough of AI in clinical practice is hampered by the black box problem. The black box problem refers to limits in the interpretability of results and to limits in explanatory functionality. Addressing the black box problem has become a major focus of research [1]. This talk describes recent attempts to addressing the black box problem in AI, offers a discussion on the suitability of those attempts for applications to oncology, and provides some future directions.

X-ray microbeams are a potential, novel mode of radiation therapy and dosimetry methods are under development that require micrometric spatial precision. The microDiamond detector has the requisite resolution and is composed of diamond which is closely tissue-equivalent. The high density of diamond however perturbs of secondary electrons and Monte Carlo methods are needed to determine corrections to accurately measure clinical parameters. The PENELOPE Monte Carlo code has been used to calculate corrections for the output factor (OF) and peak-to-valley dose ratio (PVDR). A high-performance computing (HPC) system was found to be necessary and the calculation took 72 hours when performed on a cluster of 100 CPUs. The correction for the output factor was found to be 1.009±0.016 (2 s.d.). The correction factor for the peak-to-valley ratio was found to be 1.144±0.013 (2 s.d.) and was larger due to Compton scattering of the microbeam in the extracameral components of the detector, in particular the 300 micron bulk diamond crystal. It was found that considerable improvements in efficiency could be achieved without loss of precision by switching off electron transport for electrons that are generated far from the sensitive element of the detector.

To date, more than 11,000 cancer patients have been treated with therapeutic carbon-ion beams at the National Institute of Radiological Sciences (NIRS). In the treatment planning system, the biological effectiveness of the therapeutic carbon-ion beams has been predicted based on the microdosimetric kinetic (MK) model. The MK model has a variety of applications. For instance, it can be used to predict the biological effectiveness of therapeutic carbon-ion beams under protracted irradiations as well as under hypoxic conditions. Recently, we have updated the MK model to a stochastic microdosimetric kinetic (SMK) model for a new research project about hypo-fractionated multi-ion radiotherapy, referred to as a "Quantum Scalpel". This report overviews advancement of the MK model in heavy-ion radiotherapy.

This paper presents a proof of concept of silicon carbide (SiC), a type of wideband semiconductor, based dosimeter for linear energy transfer (LET) distribution measurement at clinical carbon beam therapy field. SiC Schottky barrier diode (SBD), which has been designed for ionization detector, was utilized for spectroscopy of pristine 290 MeV/u ¹²C heavy ion therapy beams at Gunma University Heavy Ion Medical Centre (GHMC) facility, Japan. Measurement of LET distribution was successfully demonstrated at different depth with developed SiC SBD dosimeter. Also SiC SBD was exposed with clinical carbon beam with high intensity condition. Through CCE and deep level defect evaluation, sufficient radiation tolerance was examined for SiC SBD for futuristic usage in clinical carbons. Those results suggested that preliminary evaluation of SiC SBD was successful as energy-dispersive dosimeter for heavy ion therapy field.

Alpha radiation is a lethal form of radiation whose short range limits its use for cancer treatment. A unique intra-tumoral alpha radiation-based tumor ablation treatment termed Diffusing Alpha emitters Radiation Therapy (DaRT) was developed and tested for tumor ablation and stimulation of anti-tumor immunity. Radium-224 loaded wires (Alpha DaRT seeds) are inserted into the tumors and release by recoil short-lived alpha-emitting atoms. These atoms disperse in the tumor at least 5 mm from the source and spray it with highly destructive alpha radiation. DaRT was found to destroy solid malignant tumors experimental animals and in patients with cutaneous malignancies. Tumor destruction resulted in activation of specific antitumor immunity. DaRT provides, for the first time, an efficient method for treatment of the entire volume of solid tumors by alpha radiation, and could be used not only as a local treatment but also as a therapeutic strategy to induce strong systemic antitumor immune responses, which will eliminate residual disease and metastases in distant sites. This combined treatment modality holds significant potential for the treatment of non-resectable human cancers.

We present our experience with training and validation of a commercially available deep learning algorithm for organs at risk(OAR) auto-contouring. Computed tomography(CTs) with OARs from a cohort of 213 head and neck(H&N) patients were used for training the deep learning model. A separate cohort of 85 CTs and structure sets was used for validation. All OARs (13) were contoured by a single physician. Metrics such as the DICE similarity coefficient (DSC), Jaccard similarity coefficient (JSC), and volumetric difference (VD) were used to analyze contouring variation. Mean DSC and JSC values ranged 0.48-0.89 and 0.32-0.8, respectively, depending on OAR. A DSC value ≥0.7 indicated low inter-observer variability. In our study, all but one of the contours were above this threshold. DSC for the middle pharyngeal constrictor had the lowest value of all the contours. This may be due to the small volume of this structure. Qualitative assessment of auto-segmented structure samples confirmed the reliability of DSC by demonstrating the compatibility between the expert's evaluation and DSC values. Overall, we found that deep learning auto contouring is a useful tool to speed up the process of contouring in radiotherapy treatment planning.

Stereotactic ablative body radiotherapy (SABR) is demonstrating good local control for patients with inoperable primary renal cell carcinoma. In a previous pilot study we identified magnetic resonance imaging (MRI) early response biomarkers that correlate with later morphological changes in computed tomography (CT) images. These early functional changes in diffusion and perfusion following radiotherapy were observed on MRI and have the potential to identify non-responders who may benefit from adjuvant or salvage therapies. Here we detail the imaging protocol for an MRI sub-study of the Focal Ablative STereotactic Radiosurgery for Cancers of the Kidney (FASTRACK II) trial. A preliminary patient case demonstrates the high quality of the imaging data, with discussion of the improvements made from the pilot protocol for improved motion management and correction. We aim to validate the previously identified early response MRI biomarkers with this rich prospective multi-centre dataset.

Radiation dose is the therapeutic agent in radiotherapy where the objective is to maximise radiation dose to a target while minimising the dose to surrounding healthy tissues. Dose in this context is typically associated with the quantity "absorbed dose" as energy deposited per unit mass and measured in J/kg of tissue. However, even if high doses are delivered (no stochastic distribution considered) and photon or electron radiation is considered (no neutrons or heavy charged particles), there will be differences in the actual dose delivered to different tissue types as the stopping power for the electrons that deliver the vast majority of dose varies with elemental composition. Historically, radiation beam calibration and dose calculations were performed in water as a readily available, easily standardised material that closely matches the radiation properties of many human tissues. However, many superior dose calculation algorithms that have recently become available due to improved computer power (Monte Carlo Calculations, Acuros) calculate dose as deposited in the medium. The present paper examines arguments for both and proposes that based on the current scientific and political developments specification of dose as dose to medium would be the more robust and future proof choice.

Boron neutron capture therapy (BNCT) is a radiation therapy that combines neutrons and boron drugs. An industry-academia-government collaboration team is currently developing the linac-based treatment device and several peripheral devices to establish this therapy. In the project, a demonstration device, "iBNCT001", for a linac-based neutron source applicable to BNCT treatment is being developed. We are developing Tsukuba Plan, a multimodal Monte Carlo-based treatment planning system, in addition to the linac-based neutron source device. Tsukuba-Plan has adopted a Particle and Heavy Ion Transport code System (PHITS) as the Monte Carlo-based dose calculation engine. PHITS has a microdosimetric function that can compute a lineal energy distribution and determine relative biological effectiveness through combination with the stochastic microdosimetric kinetic model parameters. By utilising these functions, Tsukuba-Plan is expected to estimate the weighted doses administered to the tumour region and normal tissues in the irradiation field more accurately.

The reference adult male and female voxel phantoms described in the International Commission on Radiological Protection (ICRP) publication 110 have been successfully implemented in a Geant4 application named ICRP110Phantoms. The application allows users to simulate either the whole or a partial phantom, including as little as a single cross-sectional slice. The Geant4 application allows users to estimate the absorbed dose in individual voxels and in entire organs. As example of application, the ICRP110Phantoms was used to estimate the dose deposited by a mono-energetic 125 MeV proton pencil beam, incident on the left breast and passing through the lungs and heart, modelled in partial chest phantoms of both male and female ICRP110 phantoms. The ICRP110Phantoms will be released in Geant4 as an Advanced Example to allow its use in the wider scientific community. This Geant4 Advanced Example application can be utilised for dosimetric studies in radiotherapy, nuclear medicine and radiation protection.

In this study, the survival fraction of pancreatic cancer cells exposed to a spread-out Bragg peak (SOBP) helium-ion beam are estimated using the microdosimetric method with the microdosimetric kinetic (MK) model, by measuring the specific energy with a microdosimeter. To measure the microdosimetric spectra, a 3D mushroom microdosimeter was used by mounting it on silicon-on-insulator (SOI) substrates. At different positions of the Bragg curve of a pristine helium-ion beam of 166 MeV/u, microdosimetric spectra were measured via a scanning beam port in the National Institute of Radiological Sciences. The MK parameters were determined such that the survival fraction (SF) calculated by the MK model predicts the previously reported in vitro data. For a cuboid target of 10×10×6 cm3, a treatment plan that utilised helium-ion beam was designed from the in-house treatment planning software (TPS) to achieve a 10% SF of pancreatic cancer cells throughout the target. The physical doses and microdosimetric spectra were measured for different depths by irradiating the scanning-SOBP helium-ion beam; consequently, the SF at each position of the SOBP was predicted. The predicted SFs from measured physical dose and microdosimetric spectra were in good agreement with the planned SF from TPS.

It is recognized today that the observable radiobiological effects of ionizing radiations are strongly correlated to the clustering of damages in micrometer- and nanometer-sized subcellular structures, hence to the particle track structure. The characteristic properties of track structure are directly measurable nowadays with bulky experimental apparatuses, which cannot be easily operated in a clinical environment. It is therefore interesting to investigate the feasibility of new portable detectors able to characterize the real therapeutic beams. With this in mind, a novel avalanche-confinement Tissue Equivalent Proportional Counter (TEPC) was constructed for simulating nanometric sites down to 25 nm. Experimental cluster size distributions measured with this TEPC were compared with Monte Carlo simulations of the same experiment and with cluster size distributions measured with the Startrack nanodosimeter.

The goal of this work was to assess small-field output factors (OPF) on a newly commissioned linear accelerator (linac) using a 'correction-less' 2D monolithic array of diodes, the Duo, which has a spatial resolution of 0.2 mm. The results would validate a set of OPF extracted from the golden beam data (GBD) used to represent the dosimetric characteristics of that linac, an Elekta Versa HD (Elekta, Crawley), fit with an Agility multileaf collimator (MLC). The Duo acquired relative OPF in real time for square fields of nominal side 1, 2, 3 and 4 cm, for 6 MV with flattening filter (WFF) and 6 MV flattening filter free (FFF) photon energies. Results revealed at most a 1.0% difference in OPF when compared to baseline, and bolstered confidence in the acceptance and commissioning of the linac using local GBD as a baseline match.

Using the Microdosimetric d(z) Model in combination with PHITS-simulated specific energy probability density distributions, the relative efficiency of ⁷LiF:Mg,Ti (MTS) and ⁷LiF:Mg,Cu,P (MCP-7) thermoluminescent detectors was assessed as function of the incident energy for electrons and positrons spanning from 2 keV to 1 GeV. Additionally, the effect of the dopant concentration on the determined efficiency values was carefully investigated. Finally, the results are presented in combination with calculated specific energy frequency mean values and possible correlations were discussed.

Alanine chips exposed to high doses of radiation produces long lived free radicals that could be easily measured with electron paramagnetic resonance (EPR) spectrometers. In this study, the feasibility of using alanine dosimeters for performing rapid quality assurance of Leksell Gamma Knife (LGK) treatment plans was demonstrated. A 3D printed grid was placed inside the LGK spherical solid water phantom (SWP) for measurement of doses at isocentre and off-axis points. The EPR spectroscopy was performed on a Magnettech MS-5000 EPR/ESR spectrometer. A set of dose calibration curves were established prior to the use of alanine chips for LGK dosimetry. Absolute dose, transit dose and dose/timer linearity were performed with the alanine chips positioned at the centre of the LGK solid water phantom (SWP). Five patients of different sites were selected, and patient specific quality assurance (PSQA) was performed in the LGK SWP. The absolute dose measured with the EPR alanine dosimeter agreed well within 2% of the ion chamber results and PSQA results were within 2.1%. Alanine-based EPR dosimetry offers rapid dose measurement with high accuracy and can also be used as a dosimeter for Gamma Knife PSQA.

Volumetric repainting is considered as one of the techniques for motion mitigation in proton therapy. Faster layer switching time to deliver a volumetric repainting proton plan is very critical to reduce the overall treatment time. Recently, IBA (proton therapy vendor at the Miami Cancer Institute) has implemented a "field regulation" – a new feature to reduce the switching time between layers by applying a magnetic field setpoint to specific groups of magnets. In order to investigate the impact of field regulation and volumetric repainting technique on the spot size, several spot maps were generated. The spot sizes were measured at the isocenter and four off-axis points using the Lynx 2D scintillation detector. The average difference in spot size between two delivery sequences ("down" vs. "up" directions) for given energy at all five locations was 0.6±0.5%. The measurement results from the current study demonstrated that the impact of field regulation on the spot size was very minimal, and this was true for both the volumetric and non-volumetric techniques on a ProteusPLUS proton system with a PBS dedicated nozzle.

Recent results from pre-clinical studies investigating the so-called FLASH effect suggest that the ultrahigh pulse dose rates (UHPDR) of this modality reduces normal tissue damage whilst preserving tumour response, when compared with conventional radiotherapy (RT). FLASH-RT is characterized by average dose rates of dozens of Gy/s instead of only a few Gy/min. For some studies, dose rates exceeding hundreds of Gy/s have been used for investigating the tissue response. Moreover, depending on the source of radiation, pulsed beams can be used with low repetition rate and large doses per pulse. Accurate dosimetry of high dose-rate particle beams is challenging and requires the development of novel dosimetric approaches, complementary to the ones used for conventional radiotherapy. The European Joint Research Project "UHDpulse" will develop a measurement framework, encompassing reference standards traceable to SI units and validated reference methods for dose measurements with UHPDR beams. In this paper, the UHDpulse project will be presented, discussing the dosimetric challenges and showing some first results obtained in experimental campaigns with pulsed electron beams and laser-driven proton beams.

Recent reports have shown both poly (methyl methacrylate) (PMMA) and silica optical fibres to be ionization quenching free, making them possibly very useful dosimeters for proton beams. In this study, the response from PMMA and silica optical fibres to therapeutic proton beams are evaluated. The light output was recorded from both optical fibres, exposed to varying dose-rates of 0.5 Gy/min to 20 Gy/min from a 235 MeV isochronous cyclotron. The PMMA optical fibre was observed to have a linear dose-rate response, and a constant light emission for a constant dose-rate exposure. However, in the case of the silica optical fibres, the light output was observed to increase during a constant dose-rate exposure. If uncorrected, this accumulated dose sensitivity observed in the silica optical fibres can result in erroneous measurements.

Microdosimetry measures the stochastics of imparted energy at the micrometre scale, and is a reliable experimental technique to monitor complex radiation fields such as those used in hadron therapy. At the Legnaro National Laboratories of INFN, miniaturized gas-based microdosimeters were developed specifically for this kind of applications. However, their use outside research facilities has been hindered by the encumbrance of the gas-flow system which is used to preserve gas purity and of the high-resolution analog electronic chain. To overcome this drawback, a new detector designed to work without gas flow was developed recently. The stability and reproducibility of its response in sealed conditions were studied in two measuring shifts one year apart from each other, both with the analog electronic chain and with a compact digital acquisition system. Preliminary results confirm the possibility to operate the detector with a very compact experimental setup, which could be a major advantage in clinical facilities.

Diffusing alpha-emitters radiation therapy (DaRT) is a revolutionary brachytherapy technique used to treat solid tumours. Implant seeds are coated with ²²⁴Ra which, along its shortlived daughter atoms, emits alpha particles of high linear energy transfer (LET) and of high relative biological efficiency (RBE), creating a tumour-killing dose distribution a few mm wide. Those alpha particles are of energy between 5.67 and 8.78 MeV. DaRT is under investigation in clinical trials, but there currently is no obvious solution for dosimetry aimed at quality assurance of treatment. This study introduces alpha-RAD, a dosimeter based on a metal-oxide-semiconductor (MOS) sensor technology. Alpha-RAD was characterized with ²⁴¹Am, which emits alpha particles of energy 5.49 MeV. The results showed that alpha-RAD had good linearity with dose, with the signal increasing linearly in the range from 0 to 6.84 Gy. Also, an external bias in the range between 15 and 60 V, applied on the gate of alpha-RAD during irradiation, would optimize sensitivity to alpha particles of energies typical of DaRT. Alpha-RAD, owing to its compactness, can fit into a brachytherapy needle, to be placed next to ²²⁴Ra seed implants in the tumour, for real-time in vivo dosimetry.

Positron emission tomography (PET) is a practical tool for range verification of hadron therapy. As well, the quantitative washout of the positron emitters has a potential usefulness as a diagnostic index, but the modelling for this has not been established. In this study, we measured washout rates of rabbit brain and performed kinetic analysis to explore the washout mechanism. Six rabbit brains were irradiated by ¹¹C and ¹⁵O ion beams, and dynamic PET scan was performed using our original depth of interest (DOI)-PET prototype. The washout rate was obtained based on the two-compartment model, where efflux from tissue to blood (k₂), influx (k₃) and efflux (k₄) from the first to second compartments in tissue were evaluated. The observed k₂, k₃ and k₄ of ¹¹C were 0.086, 0.137 and 0.007 min⁻¹, and those of ¹⁵O were 0.502, 0.360 and 0.007 min⁻¹, respectively. It was suggested permeability of a molecule containing ¹¹C atoms might be regulated by a transporter. The k₂ of ¹⁵O was comparable with ¹⁵O-water. This study provides basic data for modelling of the washout effect.

We assessed the dosimetric safety of a potential diapeutic Gadolinium-based contrast agent for diagnostic MR imaging and MR-guided radiotherapy by calculating depth-dependent reactions rates in a spread-out Bragg peak (SOBP) in water. Energy-dependent cross-sections for inelastic proton reactions on Gadolinium were folded with proton energy spectra at depth. Particle transport, dose, and phase space scoring was performed using Geant4-based TOPAS Monte Carlo software. The isotopic Terbium yield at depth was calculated to be between 10³ to 10⁵ atoms/mmol_natGd per Gray SOBP dose in a 2.4 cm³ volume. At currently achievable Gadolinium concentrations of 0.2 μmol/cm³ this yields approximately 26 atoms/(cm³ Gy_SOBP), corresponding to less than 1 Bq activity. Additional dose from evaporation neutrons and subsequent capture gammas is on the order of 1 μSv/Gy_SOBP. Thus, use of this long-lived contrast agent is dosimetrically safe at current concentrations and induced radioactivity is negligible.

Intensity modulated radiotherapy is a widely used technique for accurately targeting cancerous tumours in difficult locations. As treatments are becoming more complex, new methods need to be developed to monitor them. Monolithic active pixel sensors are a viable candidate for providing upstream beam monitoring during treatment. A MAPS based system can be made thin enough to have less than 1% attenuation. We have already demonstrated leaf position resolutions below 130µm at the iso-centre for 5mm wide leaves sampled 34 times per second. We have shown that the signal due to therapeutic photons can be determined and thus the dose in patient. Furthermore, the sensor works well inside an MR-linac, allowing leaf position verification even in that challenging environment.

The University of Torino (UniTO) and the National Institute for Nuclear Physics (INFN-TO) are investigating the use of Ultra Fast Silicon Detectors (UFSD) for beam monitoring in radiobiological experiments with therapeutic proton beams. The single particle identification approach of solid state detectors aims at increasing the sensitivity and reducing the response time of the conventional monitoring devices, based on gas detectors. Two prototype systems are being developed to count the number of beam particles and to measure the beam energy with time-of-flight (ToF) techniques. The clinically driven precision (< 1%) in the number of particles delivered and the uncertainty < 1 mm in the depth of penetration (range) in radiobiological experiments (up to 10⁸ protons/s fluxes) are the goals to be pursued. The future translation into clinics would allow the implementation of faster and more accurate treatment modalities, nowadays prevented by the limits of state-of-the-art beam monitors. The experimental results performed with clinical proton beams at CNAO (Centro Nazionale di Adroterapia Oncologica, Pavia) and CPT (Centro di Protonterapia, Trento) showed a counting inefficiency <2% up to 100 MHz/cm², and a deviation of few hundreds of keV of measured beam energies with respect to nominal ones. The progresses of the project are reported.

Although MRI and SPECT are routine tools in medical imaging individually, the combination of these two modalities in multimodal imaging is still unexplored. By hybridizing high-resolution anatomical with functional imaging, combined MRI-SPECT could be a valuable addition to existing medical imaging methods, reducing overall scanning time and image co-registration errors. In contrast to PET, SPECT offers a cost-efficient range of applicable radioisotopes. This proof-of-principle study shows a modified experimental MRI-SPECT insert system consisting of an MR-compatible SPECT unit with hybrid semiconductor detectors Timepix. The insert system is equipped with CdTe pixelated sensors, tungsten collimators and a radiofrequency coil (RF). To increase the number of projections acquired, the setup is also equipped with an electromagnetic step motor Microcon SX 16-0503 for a sample rotation. Measurements were performed our own invented multimodal (¹H/^99mTc) inhomogeneous phantom filled with a liquid ^99mTc radiotracer, emitting 140.5 keV γ-rays inside the RF coil by the Bruker BioSpec 47/20 (4.7 T) MR animal scanner. For the evaluation of measured data, SPECT-back projection software was developed in Matlab. Our results pave the way for a preclinical MRI-SPECT insert system. This research was performed in the framework of the Medipix Collaboration.