As part of the physics program of the heavy-ion therapy project at GSI (Germany), we have developed a one-dimensional model and a computer code for calculating the depth-dose, dose average LET, fluence and energy disributions for protons or light ions in medium. Measurements of the depth-dose disributions for beams of 12C (195,270 and 330 MeV/u),18O (200,300 and 400MeV/u) and 20Ne (670MeV/u) were performed at the heavy-ion synchrotron SIS using water as a tissue-equivalent target. At the RIKEN ring cyclotron depth-dose and dose average LET disributions were measured for 12C and 20Ne at 135 MeV/u. The calculated distributions were compared to the measured ones and good agreement was obtained for all systems studied.
Most measurements on observers' ratings anything in behavioral sciences always involve measurement errors. Similarly, observers' scores on their ratings of medical images contain measurement errors. As it is impossible to remove these errors completely, we are forced to use unreliable data. Therefore, the practical issue is how to assess and utilize the assessed results. Generalizability study, which has been hardly discussed in the visual evaluation o n medical images, is carried out on two experiments to evaluate medical images. So far, how to determine the number of interests, subjects and observers has been frustrating investigators, however, the procedure based on the result derived from this study will become useful for the investigators in the future.
The irradiation system of a heavy charged- particle radiotherapy including a proton beam should make uniform irradiation fields for target area. In order to reduce radiation dose to normal tissues, it is required to cut sharply the uniform field by a collimator. A penumbra size of the field should be as small as possible. In order to optimize the irradiation system from this point of the view, beam dynamics passing through beam modifying devices of the irradiation system should be analyzed. Analytical function, which describes statistical characteristics of the beam transversing through the irradiation system, is obtained and applied to calculate the penumbra size of the collimated beam. The analytical functions are useful for the beam wobbling system and also for the double scatterer system which are considered to be standard irradiation methods for making uniform large irradiation fields. Analytical equations for the wobbler and dual ring double scatterer methods are given and coded. From these calculations, we found that penumbra size of the irradiation system using dual-ring double scatterer was at zero depth in a water phantom twice larger than that using the wobbler method. And the difference of penumbra was kept in the deeper region of the phantom.
In order to optimize the design of a beam-delivery system in heavy charged-particle therapy, we have developed an analytical method to calculate the sizes of the penumbra and field widths of a collimated beam. Such an analytical method is effective for optimization, and is earlier to calculate than using the Monte Carlo method. In the theoretical calculation, an exact distribution function was used, which took into account multiple-scattering and field broadening due to beam wobbling. The multiple scattering in the beam-delivery system was calculated using the Fermi-Eyges distribution function. In order to include the influence of field broadening due to beam wobbling, the Fermi-Eyges distribution function was integrated along the trajectory of the wobbled beam. The average and variance of the beam angle of the exact distribution function at the collimator were derived in order to calculate the penumbra and field width for the collimated beam. A simple formula for the penumbra size and field width was also deduced by assuming a uniform field, except for a peripheral region at the irradiation site. The calculated results by both an exact method and a simplified one were compared with the experimental results using therapeutic carbon beams with an energy per nucleon of 290,350 and 400 MeV. These results agreed well for almost all cases of the HIMAC beam.
The accuracy of Boag's formula and two-voltage technique for determining the ion collection efficiency of an ionization chamber exposed to high-intensity pulsed radiation is examined. Results with these methods were compared to the experimental ion collection efficiencies estimated with the saturation charge by graphical extrapolation. When the influence of the space charge can be ignored, results with the two-voltage technique show much better agreement with the experimental values. Results with Boag's formula give a discrepancy between the experimental values at high charge density because an unreliable value (nominal value) is used for the effective electrode spacing of the ionization chamber. Therefore, we investigated the effect of the effective electrode spacing for real ionization chamber estimated with the two-voltage technique. We found that the results with Boag's formula using the estimated effective electrode spacing agree much better with the experimental values even at high charge density.
The physical properties of screen film systems for mammography were evaluated by measuring characteristic curves, Wiener spectra and modulation transfer functions at constant developing temperature (34°). Furthermore, the developing temperature dependencies of physical properties on films for mammography were evaluated by measuring characteristic curves and noise RMS in the developing temperature extent from 26° to 44°. The mutual comparison of each corporation was not carried out; therefore it was not concluded which is the best system and the best film. In the result, it has been confirmed, which is very consequencial, that a system should be selected by an examination; for example: screening or close examination, because a relationship of trade -off is between sensitivity and granularity and sharpness. Furthermore, as the films have the higher limit of developing temperature, it has been established, which is very important, that the developing temperature must be decided by an estimate of developing temperature dependency of physical properties on films for mammography, and at that time the automatic processor of each institution must be used because the physical properties of films depend upon developer and developing time.
Six prognostic factors of tumor status of the uterine cervix at the first examination: FIGO classification, the growth type of tumor, the size of tumor at portio, the size of uterus (including cervix and body), the tumor infiltration to parametrium and the tumor infiltration to vaginal wall, were analyzed on the basis of the hazard model. Each prognostic factor was classified into four to six categories. Instead of covariate and parameter, a prognostic index which indicates the degree of the effect on the prognosis for category was applied to the hazard model. The category is arranged with the influence order of the effect on the prognosis (IOEP) according to comparison among the prognostic indexes. Nine hundred and eighty nine primary cases treated with irradiation alone at the National Institute of Radiological Sciences were applied to this study. The FIGO classification and the size of the uterus showed a great influence on the prognosis. On the other hand, the tumor infiltration to vaginal wall brought the smallest influence. The IOEP based on the prognostic index agreed with the clinical observation except a few categories of the tumor infiltration to vaginal wall.
This technical report describes a 3D dose calculation code which we have developed for proton therapy treatment planning. In order to achieve fast and accurate calculation in inhomogeneous matter like a human body, the code employs the pencil beam algorithm with the infinite-slab-layer approximation. The code handles the given configuration of proton beam, beam modifying devices, and voxel-based structure of a patient and calculates the 3D dose distribution in the patient. The code showed good performance for treatment planning in terms of speed and accuracy in comparison with detailed Monte Carlo calculations.
Our previous studies have demonstrated the finite-defference time-domain method (FD-TD method) widely used for analyzing electromagnetic field are also useful for analyzing the ultrasonic propagation in the human body. However, the conditions of absorbing boundary for the outer circumference of calculation area have not been examined at all. In this study, the application of FD -TD method was evaluated in 2-dimesional problems. Berenger's perfectly matched layer absorbing boundary theory, which has been demonstrated to be applicable for FD-TD analysis of electromagnetic field was also found able to be used without any modification as the absorbing boundary in ultrasonic FD-TD analysis.