A Novel Approach for Proton Therapy Pencil Beam Scanning Patient Specific Quality Assurance Using an Integrated Detector System and 3D Dose Reconstruction

Joseph J. Bateman^1*Sonia Escribano-Rodriguez¹Samuel Flynn^2,3Tony Price³Raffaella Radogna⁴Saad Shaikh¹Harry Barnett¹Connor Godden¹Febian Febian¹Matthew Warren¹Catherine Burne³Alison Warry⁵Lee Harrison-Carey⁵Colin Baker⁵Simon Jolly¹

• ¹University College London, London, United Kingdom
• ²National Physical Laboratory, Teddington, United Kingdom
• ³School of Physics and Astronomy, University of Birmingham, Birmingham, United Kingdom
• ⁴University of Bari Aldo Moro, Bari, Italy
• ⁵University College London Hospital, London, United Kingdom

Current proton beam therapy patient-specific quality assurance (PSQA) methods rely on time-intensive phantom measurements or machine-reported parameters without independent verification. This work presents an integrated detector system for phantom-less Pencil Beam Scanning (PBS) PSQA, providing independent spot-by-spot measurements of all critical beam parameters and 3D dose reconstruction. The integrated detector combines three separate systems: a scintillator range telescope for range and energy measurement; a CMOS pixel sensor for spot position and size verification; and a Transmission Calorimeter (TC) for beam intensity measurements. Measured parameters feed Monte Carlo simulations to reconstruct 3D dose distributions for comparison with treatment planning predictions. Validation was performed at UCLH using single spot position spread out Bragg Peak (SOBP) and 5×5×10 spot box field configurations. Energy values obtained from range measurements showed strong correlation with DICOM values (R^2 > 0.998) with an accuracy of between 2.17 mm and 1.23 mm for different beam deliveries. CMOS pixel sensor measurements succeeded for single spot fields but experienced saturation at higher intensities and incomplete coverage for the larger box field. The TC demonstrated excellent dose linearity (R2 = 1.000). Monte Carlo reconstructions agreed well with reference simulations for longitudinal profiles, though lateral reconstructions proved challenging with 77% gamma pass rates (2%/2mm) for the box field. This proof-of-concept demonstrates feasibility of independent beam parameter verification for PBS PSQA while maintaining patient geometry. The approach offers advantages over current methods but requires resolution of energy calibration offsets and detector limitations before clinical implementation. Future work will address these challenges and expand validation to clinical treatment plans.