|
内容記述 |
In-beam PET is a practical technique for verifying the beam range in particle therapy, as nuclear fragmentation reactions produce positron emitters along the beam path. Since these emitters are generated in small quantities and have short half-lives, data must be collected during or immediately after the irradiation. This in-beam or online measurement also helps minimize patient restraint time. To enable 3D in-beam PET, we developed OpenPET, a PET system based on a full-ring geometry in which the edges of a cylinder are cut by two aslant planes to allow the treatment beam to pass through. A human-scale OpenPET system has been installed in a treatment room of the Heavy Ion Medical Accelerator in Chiba (HIMAC), and the first patient image for head and neck cancer treatment was reported at the last IEEE MIC. Since then, additional patient cases have been accumulated, enabling inter-fraction comparison. Each fraction included both vertical and horizontal irradiations; since the current OpenPET system supports only horizontal irradiation, PET acquisition was performed for the horizontal irradiation only, starting at beam-on and continuing for 10 min after the irradiation. In this study, we evaluated the inter-fraction consistency of in-beam PET images across treatment fractions delivered under the same treatment plan using gamma analysis, a metric that compares two images based on spatial and intensity agreement. As a result, the gamma index evaluation showed high pass rates, indicating no significant variation in in-beam PET images across fractions. Anatomical changes during the treatment course, such as tumor shrinkage or nasal condition, can affect the beam range and may increase the dose to normal tissues, changing positron emitter distributions. Therefore, the results with the consistent in-beam PET images showed the beam range stability, supporting the utility of OpenPET for quality assurance in carbon-ion therapy. |
