GATE Monte Carlo Simulation of the RS 3400 X-ray Blood Irradiator: Dose Distribution Analysis in a MAGIC-f Gel Blood-Bag Phantom

Abstract

Accurate spatial dose verification is essential for ensuring the reliability of compact X-ray blood irradiators used to prevent transfusion-associated graft-versus-host disease. This work develops and characterizes a Monte Carlo dosimetric model of the RS 3400 self-contained X-ray blood irradiator operating at 150 kV / 10 mA for a 280 s protocol, using the GATE v9.0 toolkit built on Geant4. The model reproduces the gold-anode 150 kV X-ray spectrum, a seven-component lead-shielding assembly, an ellipsoidal MAGIC-f polymer-gel blood-bag phantom, and the combined self-rotation (6°/s) and orbital revolution (3°/s) of the sample, discretized into 28 000 time slices over the irradiation period with 10¹⁰ primary photons tracked. A unified MATLAB post-processing pipeline converts the energy-deposition output to absorbed dose and performs voxel-wise uncertainty, regional, inner-subregion, and ROI-based analyses in the coronal plane. The simulated distribution exhibits a homogeneous central plateau (inner-region mean 23.92 Gy, uniformity ratio 1.08) that decreases radially toward the periphery, with a mean voxel-wise statistical uncertainty of 0.771 %. The results confirm that the combined motion model effectively homogenizes the central dose and that the model is converged to a clinically useful precision.