DOSIS: a patient-specific MC based dosimetry toolkit for nuclear medicine procedures

P. PÉREZ

Viena

Simposio; International Symposium on Standards, Applications and Quality Assurance in Medical Radiation Dosimetry - IDOS 2019; 2019

International Atomic Energy Agency

The use of radiolabelled molecules for tumor targeting proved to be very useful for the treatment of systemic malignancies, even when external radiotherapy is not applicable or appropriate. To this aim, radionuclides are labelled to carrier molecules for transporting them to the tumor region after patient infusion.In order to achieve lethal damage to tumor cells and try to avoid exceeding tolerable dose levels in normal tissues, an accurate 3D patient-specific dosimetry assessment with millimeter resolution should be a relevant pre-requisite for these nuclear medicine procedures. Hereby, the development of dosimetry tools to asses in vivo radiopharmaceutical biodistribution for further estimation of 3D dose released to target and normal tissues has become in an increasing research line in the field of internal dosimetry.The three most extended methods for calculating relevant dosimetry parameters in nuclear medicine are: S-values estimation, Dose Point Kernel (DPK) convolution and Monte Carlo (MC) simulation of radiation transport and computation of absorbed energy. These three methods have shown to be capable of dosimetry at voxel level, which is in fact desired in order to consider the non homogeneity for both activity and mass distributions based on patient specific images.S-values are based on virtual mathematical phantoms calculations and some corrections based on patient characteristics, DPK convolution is able to achieve radial dose distributions in patient-specific conditions from activity distribution acquired from patient images (i.e. SPECT or PET) and MC simulations are accepted as the most accurate method of estimating radiation transport and absorbed dose in complex geometries.Then, it is necessary to advance towards integrated solutions capable of performing patient- specific 3D dosimetry at voxel level based on anatomical and metabolic images by using at least one of the two last mentioned methods: DPK convolution and/or MC simulations; especially in a context of growth of the use of radionuclides for therapy and the energies involved.