An open-source framework for image-based planning and PET/computed tomography dosimetry in yttrium-90 radioembolization

The aim of this study was to describe Yttrium Radioembolization Dosimetry Evaluation (YRDose), an investigational open-source software suite for image-based activity planning and optional post-therapy dosimetry in yttrium-90 (Y-90) radioembolization. YRDose was implemented in Python as scripted modules within 3D Slicer 5.8.1. Mandatory inputs for the complete pre-therapy workflow are anatomical computed tomography (CT) and quantitative technetium-99m macroaggregated albumin single-photon emission computed tomography/CT image. Quantitative Y-90 PET/CT is optional for delivered-dose verification. The suite integrates CT-based segmentation with expert review, geometric alignment, binary-mask export, three-dimensional lung shunt fraction (LSF) calculation, compartment-based activity allocation, and mean absorbed-dose estimation using the Medical Internal Radiation Dose local energy deposition formalism. Segmentations, intermediate measurements, and final outputs are stored in a single Medical Reality Markup Language scene. The framework enabled sequential segmentation, volumetric LSF calculation, compartment mass estimation, planned activity calculation, and post-therapy mean absorbed-dose verification within the same software environment, preserving traceability and reproducibility. This technical note presents a reproducible open-source platform for Y-90 radioembolization planning and PET/CT dosimetry. YRDose provides an accessible research framework for developing, testing, and auditing image-based personalized dosimetry workflows, particularly in institutions without proprietary platforms. Formal validation remains necessary before routine clinical adoption.