Implementation and validation of collapsed cone superposition for dosimetry of radionuclides with photon component

105 Objectives A collapsed cone superposition (CCS) has been recently implemented for absorbed dose (AD) calculation of beta emitters (Sanchez-Garcia PMB 2014). This paper presents the implementation for radionuclides with a photon component. Methods The CCS relies in a water dose kernel (DK) scaled to take into account tissue density heterogeneities and approximates the energy transport to a finite number of directions to reduce computation complexity. CCS was extended and validated on 18F, 99mTc, 131I and 177Lu. DKs were generated with MCNP (sampling 0.5 mm). The validation was done towards MCNP in 6 phantoms constituted of 2 concentric spheres (radii of 4 and 8 cm) representing interfaces between soft-tissue, lung and bone. The AD was scored in concentric shells of 1 mm leading to a radial AD plot. The metrics used for comparison were the percentage of points in the plot satisfying the γ(3%, 3 mm) and γ(5%, 5 mm) criterions. Results For 18F, γ(3%, 3 mm) was 100% for all phantoms. For 99mTc, γ(3%, 3 mm) was between 71 and 97%, except for the bone/lung interface (7%), and γ(5%, 5 mm) was over 99% except for the bone/lung phantom too (57%). For 131I, γ(3%, 3 mm) was between 99 and 100%, except for the bone/lung phantom (92%), but γ(5%, 5 mm) was 100% for all phantoms. For 177Lu, γ(3%, 3 mm) was between 89 and 100%, except for bone/lung phantom (59%), but γ(5%, 5 mm) was over 97% for all phantoms. Calculation times were around 4 minutes on a 3.2GHz Intel Core i5 machine with 4 cores. Conclusions Our results show that the CCS algorithm has a good agreement with MC simulations for various tissue interfaces and common radionuclides, confirming that it is a reliable alternative in radiopharmaceutical dosimetry. Research Support INSERM grant n°PC201227