BEAM3R, a dose estimation framework operating in beam's-eye-view (BEV), combines a Mamba-3 state-space depth-sequence core with physics-based transport conditioning to model long-range depth transport without expensive 3D convolutions.
Abstract
To enable accurate and rapid photon control point and proton beamlet dose calculation in the DoseRAD2026 challenge, we present BEAM3R, a dose estimation framework operating in beam's-eye-view (BEV). Our core innovation combines a Mamba-3 state-space depth-sequence core with physics-based transport conditioning to model long-range depth transport without expensive 3D convolutions. BEAM3R shares a 2D CNN encoder-decoder architecture for photon and proton dose tasks, processing per-plane BEV slices. Proton beamlets are conditioned on water equivalent thickness and remaining range, encoding the parameters determining Bragg peak position. Photon models use a bidirectional Mamba-3 core to capture dose contributions from materials downstream of the calculation point, while the proton model uses a forward core with learned energy-prefix tokens and a Bragg-peak refinement module. To reduce interpolation artifacts and support high spatial resolution, we introduce axial grid alignment of BEV lattices with CT slices and an implicit super-resolution representation via sub-pixel phase packing, evaluated by a differentiable Triton-accelerated resampler that reconstructs packed cubic B-spline coefficients directly in CT space. For MRI-based tasks, synthetic CTs (sCT) are generated by a patch-based conditional GAN with a SwinUNETR backbone. On the preliminary DoseRAD2026 test set, CT-to-photon and CT-to-proton models achieved 1%/1 mm local gamma pass rates of 96.8% and 96.0%, with stratified plan-level MAEs of 0.0041 and 0.0079. Substituting sCT reduced gamma pass rates to 89.7% for photon and 75.4% proton plan level doses, with stratified plan-level MAEs of 0.0093 and 0.0336. Standardised runtimes were 23.4 s and 18.4 s for CT-to-photon and CT-to-proton prediction, increasing to 39.7 s and 42.8 s for the corresponding MRI-based pipelines.
Purpose: Deep learning can accelerate radiotherapy dose prediction, but particle-therapy datasets remain scarce. We evaluated whether a physics-informed ensemble trained exclusively on photon plans could predict proton and carbon-ion dose distributions without particle-dose training.Methods: Using matRad, 90 reference...
Muhammed Emin Bedi̇r· Acibadem Universitesi Saglik...· 0 citations
Machine learning models were trained on simulated datasets generated using Geant4 and MCDE and implemented through the Big Data REU Integrated Development and Experimentation (BRIDE) platform to classify and filter scatter events for more accurate image reconstruction.
Ayan Kabaria, M. Khalid, Sophia Lopez et al.· 1 citation
Deep learning-based single-shot x-ray phase-contrast reconstruction can suppress artifacts, reduce radiation dose and mechanical positioning requirements. This study compares UNet3+, generative adversarial network (GAN), and denoising diffusion implicit model (DDIM) for phase-contrast imaging in single absorption grati...
Jia-Ming Ye, Jian-Heng Huang, Xin Liu et al.· International Conference on...· 0 citations
We present a deep-learning-based phase-retrieval framework that uses Bessel-beam illumination to extend the recoverable aberration range in wavefront sensing. By exploiting the large effective numerical aperture and extended depth of focus of Bessel beams, the proposed method increases the range of measurable aberratio...
Zhao-Wei Chen, Ryan Luu, R. Parks et al.· Optical Manufacturing and Te...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.