Yuto Nagao

Beam range estimation by measuring bremsstrahlung

We describe a new method for estimating the beam range in heavy-ion radiation therapy by measuring the ion beam bremsstrahlung. We experimentally confirm that the secondary electron bremsstrahlung process provides the dominant bremsstrahlung contribution. A Monte Carlo simulation shows that the number of background photons from annihilation gamma rays is about 1% of the bremsstrahlung strength in the low-energy region used in our estimation (63-68 keV). Agreement between the experimental results and the theoretical prediction for the characteristic shape of the bremsstrahlung spectrum validates the effectiveness of our new method in estimating the ion beam range.

Optimization and verification of image reconstruction for a Compton camera towards application as an on-line monitor for particle therapy

Particle therapy is an advanced cancer therapy that uses a feature known as the Bragg peak, in which particle beams suddenly lose their energy near the end of their range. The Bragg peak enables particle beams to damage tumors effectively. To achieve precise therapy, the demand for accurate and quantitative imaging of the beam irradiation region or dosage during therapy has increased. The most common method of particle range verification is imaging of annihilation gamma rays by positron emission tomography. Not only 511-keV gamma rays but also prompt gamma rays are generated during therapy; therefore, the Compton camera is expected to be used as an on-line monitor for particle therapy, as it can image these gamma rays in real time. Proton therapy, one of the most common particle therapies, uses a proton beam of approximately 200 MeV, which has a range of ~ 25 cm in water. As gamma rays are emitted along the path of the proton beam, quantitative evaluation of the reconstructed images of diffuse sources becomes crucial, but it is far from being fully developed for Compton camera imaging at present. In this study, we first quantitatively evaluated reconstructed Compton camera images of uniformly distributed diffuse sources, and then confirmed that our Compton camera obtained 3 %(1 σ) and 5 %(1 σ) uniformity for line and plane sources, respectively. Based on this quantitative study, we demonstrated on-line gamma imaging during proton irradiation. Through these studies, we show that the Compton camera is suitable for future use as an on-line monitor for particle therapy.

First demonstration of multi-color 3-D in vivo imaging using ultra-compact Compton camera

In the field of nuclear medicine, single photon emission tomography and positron emission tomography are the two most common techniques in molecular imaging, but the available radioactive tracers have been limited either by energy range or difficulties in production and delivery. Thus, the use of a Compton camera, which features gamma-ray imaging of arbitrary energies from a few hundred keV to more than MeV, is eagerly awaited along with potential new tracers which have never been used in current modalities. In this paper, we developed an ultra-compact Compton camera that weighs only 580 g. The camera consists of fine-pixelized Ce-doped Gd3Al2Ga3O12 scintillators coupled with multi-pixel photon counter arrays. We first investigated the 3-D imaging capability of our camera system for a diffuse source of a planar geometry, and then conducted small animal imaging as pre-clinical evaluation. For the first time, we successfully carried out the 3-D color imaging of a live mouse in just 2 h. By using tri-color gamma-ray fusion images, we confirmed that 131I, 85Sr, and 65Zn can be new tracers that concentrate in each target organ.

Imaging of monochromatic beams by measuring secondary electron bremsstrahlung for carbon-ion therapy using a pinhole x-ray camera

A feasibility study on the imaging of monochromatic carbon-ion beams for carbon-ion therapy was performed. The evaluation was based on Monte Carlo simulations and beam-irradiation experiments, using a pinhole x-ray camera, which measured secondary electron bremsstrahlung (SEB). The simulation results indicated that the trajectories of the carbon-ion beams with injection energies of 278, 249 and 218 MeV/u in a water phantom, were clearly imaged by measuring the SEB with energies from 30 to 60 keV, using a pinhole camera. The Bragg-peak positions for these three injection energies were located at the positions where the ratios of the counts of SEB acquisitions to the maximum counts were approximately 0.23, 0.26 and 0.29, respectively. Moreover, we experimentally demonstrated that it was possible to identify the Bragg-peak positons, at the positions where the ratios coincided with the simulation results. However, the estimated Bragg-peak positions for the injection energies of 278 and 249 MeV/u were slightly deeper than the expected positions. In conclusion, for both the simulations and experiments, we found that the 25 mm shifts in the Bragg-peak positions can be observed by this method.

In vivo simultaneous imaging with 99m Tc and 18 F using a Compton camera

We have been developing a medical imaging technique using a Compton camera. This study evaluates the feasibility of clear imaging with 99mTc and 18F simultaneously, and demonstrates in vivo imaging with 99mTc and/or 18F. We used a Compton camera with silicon and cadmium telluride (Si/CdTe) semiconductors. We estimated the imaging performance of the Compton camera for 141 keV and 511 keV gamma rays from 99mTc and 22Na, respectively. Next, we simultaneously imaged 99mTc and 18F point sources to evaluate the cross-talk artifacts produced by a higher energy gamma-ray background. Then, in the in vivo experiments, three rats were injected with 99mTc-dimercaptosuccinic acid and/or 18F-fluorodeoxyglucose and imaged. The Compton images were compared with PET images. The rats were euthanized, and the activities in their organs were measured using a well counter. The energy resolution and spatial resolution were measured for the sources. No apparent cross-talk artifacts were observed in the practical-activity ratio (99mTc:18F  =  1:16). We succeeded in imaging the distributions of 99mTc and 18F simultaneously, and the results were consistent with the PET images and well counter measurements. Our Si/CdTe Compton camera can thus work as a multi-tracer imager, covering various SPECT and PET probes, with less cross-talk artifacts in comparison to the conventional Anger cameras using a collimator. Our findings suggest the possibility of human trials.

Monte Carlo simulation of photon emission below a few hundred kiloelectronvolts for beam monitoring in carbon ion therapy

The purpose of this study is to determine whether the main component of the low-energy (63-68 keV) particles emitted perpendicularly to the 12C beam from the 12C-irradiated region in a water phantom is secondary electron bremsstrahlung (SEB). Monte Carlo simulations of a 12C-beam (290 MeV/u) irradiated on a water phantom were performed. A detector was placed beside the water phantom with a lead collimator between the phantom and the detector. To move the Bragg-peak position, a binary filter was placed in an upper stream of the phantom. The energy distributions of the particles incident on the detector and those deposited in the detector were analyzed. The simulation was also performed with suppressed delta-ray and/or bremsstrahlung generation to identify the SEB components. It was found that the particles incident on the detector were predominantly photons and neutrons. The yields of the photons and energy deposition decreased with the suppression of SEB generation. It is concluded that one of the predominant components of the yields in the regions shallower than the Bragg-peak position is due to SEB generation, and these components become significantly smaller in regions deeper than the Bragg-peak position.

Detection of a gas region in a human body across a therapeutic carbon beam by measuring low-energy photons

We studied feasibility of detection of a gap which is located across a beam track by measuring low-energy (63–68 keV) photons generated by beam irradiation. An experiment was performed with the Heavy Ion Medical Accelerator in Chiba (HIMAC). A 12C beam having 290 MeV/u was injected on a target consisting of two acrylic blocks. These two blocks were placed with a 10 mm gap along the beam axis. A detection system consisting of a semiconductor detector, a lead collimator having a slit, and borated polyethylene blocks was placed on a movable stage to detect low-energy photons emitted perpendicularly to the beam axis. The position of the detection system was moved at 2 mm intervals along the beam axis. It was found that the yield of 63–68 keV photons was clearly correlated with the position of the detection system. The position at which the yield curve had the lowest value agreed with the gap position. We also confirmed that the experimental result was well reproduced by a Monte Carlo simulation that includes generation of secondary electron bremsstrahlung.

A new method for monitoring beam range by measuring low energy photons

We studied a new method to monitor a beam range in heavy-ion radiation therapy by measuring low energy photons emitted from a track of the ion beam. A 290 MeV/u carbon beam was injected into a cylindrical water phantom. A CdTe semiconductor detector with a lead slit having a width of 2mm was placed at a side of the phantom. In order to measure the position dependence of the low energy photon count, the beam range was varied by changing the injection energy using a binary energy degrader placed about 100 cm upstream of the beam focal point. The measured photon count decreased when the detector got closer to the end point of the beam range and the derivative of the photon count clearly changed in front of the range position. This was explained by our theoretical study assuming the photons were secondary electron bremsstrahlung. These results indicate that this new method could estimate the range position from the observation of bremsstrahlung with an accuracy of a few mm.

An evaluation of three-dimensional imaging by use of Si/CdTe Compton cameras

Imaging technique of RI tracer for physiological function analysis is substantially useful and popular as positron emission tomography (PET), positron-emitting tracer imaging system (PETIS) or single photon emission computed tomography (SPECT). Recently, Compton camera is thought to become a promising imaging apparatus in several field, for example, medical, biological application and security inspection, because of its simultaneous imaging ability against the wide energy range gamma-rays from a few hundred keV to a few MeV. In this work, three-dimensional imaging ability of our Compton camera system developed for human body imaging has been studied by use of Monte Carlo simulation. The imaging system consists of plural Compton-camera head-modules having the diameter of 20 cm and height of 20 cm. Each of head modules has a 32 mm wide Si double-side strip detector (DSD) and four CdTe-DSDs stacked at intervals of 4 mm. The head modules are settled on XYZ-stages and can be moved by XYZ-stage around the subject imaged during data acquisition. The simulation study was performed by use of a Geant4-based Compton-camera simulator developed by JAXA/ISAS. Three sphere-shaped gamma-sources, having diameters of 10, 13 and 17 mm, of the NEMA NU2-2007 standard body-phantom was assumed to be placed around the center of simulation space. The RI density is uniform and the energy of gamma ray 511 keV. Six camera-placements around the back and a side of the body phantom. Three-dimensional image was reconstructed by use of the List-Mode Expectation-Maximizing Maximum-Likelihood method. The positions of each sphere were clearly identified at the correct positions. The RI distributions of the imaging result are not asymmetrical between X- and Z-direction, which is found to be owing to the asymmetrical placement of the detector stacks. The RI intensities deduced from the image tend to reflect the real RI intensities although improvement is still required.