A. Ishizaki

Dynamic splitting of Gaussian pencil beams in heterogeneity-correction algorithms for radiotherapy with heavy charged particles

The pencil-beam algorithm is valid only when elementary Gaussian beams are small enough compared to the lateral heterogeneity of a medium, which is not always true in actual radiotherapy with protons and ions. This work addresses a solution for the problem. We found approximate self-similarity of Gaussian distributions, with which Gaussian beams can split into narrower and deflecting daughter beams when their sizes have overreached lateral heterogeneity in the beam-transport calculation. The effectiveness was assessed in a carbon-ion beam experiment in the presence of steep range compensation, where the splitting calculation reproduced a detour effect amounting to about 10% in dose or as large as the lateral particle disequilibrium effect. The efficiency was analyzed in calculations for carbon-ion and proton radiations with a heterogeneous phantom model, where the beam splitting increased computing times by factors of 4.7 and 3.2. The present method generally improves the accuracy of the pencil-beam algorithm without severe inefficiency. It will therefore be useful for treatment planning and potentially other demanding applications.

Microbeam Analysis at Tohoku University for Biological Studies

A microbeam analysis system at Tohoku University has been improved in detection efficiency for application to single cell analysis. The system is applicable to STIM analysis and to simultaneous PIXE and RBS analysis. Sample preparation methods suitable for non-adhesive single cell analysis were developed and first results with the improved analysis system are shown.

Computational modeling of beam-customization devices for heavy-charged-particle radiotherapy

A model for beam customization with collimators and a range-compensating filter based on the phase-space theory for beam transport is presented for dose distribution calculation in the treatment planning of radiotherapy with protons and heavier ions. Independent handling of pencil beams in conventional pencil-beam algorithms causes unphysical collimator-height dependence in the middle of large fields, which is resolved by the framework comprised of generation, transport, collimation, regeneration, range-compensation and edge-sharpening processes with a matrix of pencil beams. The model was verified to be consistent with measurement and analytic estimation at a submillimeter level in the penumbra of individual collimators with a combinational-collimated carbon-ion beam. The model computation is fast, accurate and readily applicable to pencil-beam algorithms in treatment planning with the capability of combinational collimation to make the best use of the beam-customization devices.

The grid-dose-spreading algorithm for dose distribution calculation in heavy charged particle radiotherapy

A new variant of the pencil-beam (PB) algorithm for dose distribution calculation for radiotherapy with protons and heavier ions, the grid-dose spreading (GDS) algorithm, is proposed. The GDS algorithm is intrinsically faster than conventional PB algorithms due to approximations in convolution integral, where physical calculations are decoupled from simple grid-to-grid energy transfer. It was effortlessly implemented to a carbon-ion radiotherapy treatment planning system to enable realistic beam blurring in the field, which was absent with the broad-beam (BB) algorithm. For a typical prostate treatment, the slowing factor of the GDS algorithm relative to the BB algorithm was 1.4, which is a great improvement over the conventional PB algorithms with a typical slowing factor of several tens. The GDS algorithm is mathematically equivalent to the PB algorithm for horizontal and vertical coplanar beams commonly used in carbon-ion radiotherapy while dose deformation within the size of the pristine spread occurs for angled beams, which was within 3 mm for a single 150-MeV proton pencil beam of 30 degrees incidence, and needs to be assessed against the clinical requirements and tolerances in practical situations.

MICROBEAM ANALYSIS OF SINGLE AEROSOL PARTICLES AT TOHOKU UNIVERSITY

A microbeam system has been developed for the analysis of single aerosol particles. Combination of PIXE, RBS and off-axis STIM methods enabled simultaneous analysis for hydrogen to metal elements. Aerosol particles were collected on thin polycarbonate film (~0.3 μm) resulting in good signal-to-noise ratio. Quantitative elemental correlation was measured for single aerosol particles. A total of 270 particles were analyzed and clustered into 4 groups. The analysis system reveals the chemical composition of aerosol particles and is a powerful tool for source identification.

DEVELOPMENT OF AN IMAGE RECONSTRUCTION METHOD FOR MICRON-CT USING PIXE

A prototype of micron-CT for biological research is being developed at Tohoku University. This micron-CT uses a point X-ray source with a spot size of 1μm and an X-ray CCD with 1000×1000 pixels of 8μm×8μm, achieving a spatial resolutions of the order of micro-meter. The event data obtained by the X-ray CCD is statistically poor and the 3 dimensional filtered back projection (3D FBP) algorithm, generally used in image reconstruction of X-ray CT, is not suitable because it is highly sensitive to statistical noise. Hence, we applied the expectation maximization (EM) algorithm for image reconstruction and developed an image reconstruction method using 3D EM algorithm. To confirm the validity of the reconstruction method, we irradiated two hairs inside a micro tube and reconstructed the CT image applying both EM and FBP algorithm on projection data. With 200×200×200 voxels of 4μm×4μm×4μm in the field of view, the computation time was less than 2 mins per iteration on a DELL Precision 650 Workstation 3.2GHz. The resulting EM image showed a better contrast than FBP image, and in the EM reconstructed CT image, we were able to reconstruct the micro tube of 270μm diameter and 45μm wall thickness and to visualize the two hairs inside.

In-Vivo Elemental Analysis by PIXE-μ-CT

We have developed “micron-CT”, using micro-PIXE for in-vivo imaging. This system comprises an X-ray CCD camera (Hamamatsu photonics C8800X9) with high resolution (pixel size: 8×8μm2, number of pixels: 1000×1000) and an X-ray-point-source with a spot size of 1.5×1.5μm2 which is generated by irradiation of a microbeam on a pure metal target. Thus we can acquire projection data with high resolution. The sample is placed in a small diameter tube and is rotated by a stepping motor. The 3D images were reconstructed from the obtained projection data by using cone-beam CT reconstruction algorithm. X-ray spectra produced by heavy charged particle bombardment, exhibit a much smaller continuous background compared to electron bombardment. Therefore, X-rays produced by ion beam can be used as a monochromatic and low energy X-ray source. The feature is very effective to investigate small insects. Moreover we can get elemental distribution image of object by choosing appropriate characteristic X-rays corresponding to the absorption edge. On the other hand, the conventional X-ray CT, in which continuous X-rays are used, provides images of the electron density in the object. Using this system, we were able to get 3D images of a living ants head with 6 μm spatial resolution. By using Fe-K-X-rays (6.40 keV) and Co-K-X-rays (6.93 keV), we can investigate the 3D distribution of Mn (K-absorption edge = 6.54 keV) in an ants head.

Development of an In-air on/off Axis STIM System for Quantitative Elemental Mapping

We have developed an in-air on/off axis STIM for simultaneous density mapping with PIXE and RBS, which will be useful for damage-monitoring in cell analysis and for yield correction based on the thickness distribution of X-ray self-absorption in samples. The in-air on/off axis STIM system provides a mass concentration map in the cell analysis. In the system, a thin scattering foil is placed downstream of the sample and scattered protons are detected by a Si-PIN photodiode set at 30 degrees with respect to the beam axis. These components are set in a He-gas-filled chamber to reduce energy loss, scattering and sample damage. Using this system, areal density mapping is carried out for RBL-2H3 cells simultaneously with PIXE and RBS. Correction for self-absorption is performed and areal density map of elements is converted into a mass-concentration map using the measured matrix density. The areal density distribution of P corresponds to that of matrix and mass concentration of P is uniform in the cell region. On the other hand, Br is concentrated in the nucleus, even in the mass concentration map. The Br accumulation in the nucleus is first confirmed in mass concentration using the on/off axis STIM and PIXE system. The in-air on/off STIM system will be effective for monitoring changes in cell density during beam irradiation.

Development of an irradiation method with lateral modulation of SOBP width using a cone‐type filter for carbon ion beams

Passive irradiation methods deliver an extra dose to normal tissues upstream of the target tumor, while in dynamic irradiation methods, interplay effects between dynamic beam delivery and target motion induced by breathing or respiration distort the dose distributions. To solve the problems of those two irradiation methods, the authors have developed a new method that laterally modulates the spread-out Bragg peak (SOBP) width. By reducing scanning in the depth direction, they expect to reduce the interplay effects. They have examined this new irradiation method experimentally. In this system, they used a cone-type filter that consisted of 400 cones in a grid of 20 cones by 20 cones. There were five kinds of cones with different SOBP widths arranged on the frame two dimensionally to realize lateral SOBP modulation. To reduce the number of steps of cones, they used a wheel-type filter to make minipeaks. The scanning intensity was modulated for each SOBP width with a pair of scanning magnets. In this experiment, a stepwise dose distribution and spherical dose distribution of 60 mm in diameter were formed. The nonflatness of the stepwise dose distribution was 5.7% and that of the spherical dose distribution was 3.8%. A 2 mm misalignment of the cone-type filter resulted in a nonflatness of more than 5%. Lateral SOBP modulation with a cone-type filter and a scanned carbon ion beam successfully formed conformal dose distribution with nonflatness of 3.8% for the spherical case. The cone-type filter had to be set to within 1 mm accuracy to maintain nonflatness within 5%. This method will be useful to treat targets moving during breathing and targets in proximity to important organs.

SU‐FF‐T‐626: Dynamic Splitting of Gaussian Pencil Beams in Heterogeneity‐Correction Algorithms for Radiotherapy with Heavy Charged Particles

Purpose: To develop an algorithm to resolve intrinsic problems with dose calculations using pencil beams when particles involved in each beam are overreaching a lateral density interface or when they are detouring in a laterally heterogeneous medium. Method and Materials: A finding on a Gaussian distribution, such that it can be approximately decomposed into multiple narrower, shifted, and scaled ones, was applied to dynamic splitting of pencil beams implemented in a dose calculation algorithm for proton and ion beams. The method was tested in an experiment with a range‐compensated carbon‐ion beam. Its effectiveness and efficiency were evaluated for carbon‐ion and protonbeams in a heterogeneous phantom model. Results: The splitting dose calculation reproduced the detour effect observed in the experiment, which amounted to about 10% at a maximum or as large as the lateral particle‐disequilibrium effect. The proton‐beam dose generally showed large scattering effects including the overreach and detour effects. The overall computational times were 9 s and 45 s for non‐splitting and splitting carbon‐ion beams and 15 s and 66 s for non‐splitting and splitting protonbeams.Conclusions: The beam‐splitting method was developed and verified to resolve the intrinsic size limitation of the Gaussian pencil‐beam model in dose calculation algorithms. The computational speed slowed down by factor of 5, which would be tolerable for dose accuracy improvement at a maximum of 10%, in our test case.