Physics-constrained and geometry-aware deep reconstruction for limited-angle CBCT.
Authors
Affiliations (7)
Affiliations (7)
- Computer Science and Informatics, Emory University, 400 Dowman Drive, Atlanta, Georgia, 30322, United States.
- Department of Radiology Oncology, Emory University, 1365 CLIFTON RD NE, ATLANTA, Atlanta, Georgia, 30322, United States.
- Department of Radiation Oncology, Emory University, 100 Woodruff Circle, Atlanta, Atlanta, Georgia, 30322-1007, United States.
- Department of Radiology Oncology, Emory University, 1365 Clifton RD NE, Atlanta, Georgia, 30322-1007, United States.
- Department of Radiology and Sciences Imaging Department of Radiology Oncology, Emory University, 1365 Clifton RD NE, Atlanta, Georgia, 30322, United States.
- Department of Radiation & Cellular Oncology, The University of Chicago, 5841 S. Maryland Avenue, MC 0085, Chicago, Illinois, 60637, United States.
- Department of Radiology Oncology, Emory University, Clifton Rd, Atlanta, Georgia, 30322-1007, United States.
Abstract
Limited-angle cone-beam computed tomography (CBCT) is attractive for imageguided radiotherapy and imaging-guided procedure when full-arc acquisition is impractical, but the incomplete angular sampling leads to severe streaking, shading, and structural distortion. This study aimed to develop a physics-constrained and geometry-aware deep reconstruction framework for recovering anatomically faithful volumetric CBCT from limited-angle projections.
Approach: We propose PCGA-Net, a dual-domain reconstruction framework that combines projection completion, analytic geometric reconstruction, and image-domain refinement within a unified cross-domain design. Specifically, a projection completion module first estimates missingangle information from the measured limited-angle projections. A geometry-aware reconstruction module then maps the completed projections into image space using differentiable cone-beam forward and backward-projection operators. Finally, a volumetric refinement module suppresses residual artifacts and restores anatomical detail under geometry-guided cross-domain consistency supervision. The framework was evaluated on three cohorts: gynecologic highdose-rate brachytherapy, abdominal liver 4DCT, and head-and-neck CBCT.
Main Results: Across all three cohorts, PCGA-Net consistently outperformed conventional limitedangle FDK reconstruction and the compared learning-based baselines, cGAN and PatRecon-Net, in both qualitative and quantitative evaluation. On the gynecologic HDR brachytherapy cohort, PCGA-Net achieved an MAE of 34.47 ± 6.28 HU, an SSIM of 0.87 ± 0.04, and a PSNR of 35.79 ± 0.95 dB. On the abdominal liver 4DCT cohort, PCGA-Net achieved 39.27 ± 6.10 HU MAE, 0.82 ± 0.03 SSIM, and 36.85 ± 1.16 dB PSNR.On the head-and-neck CBCT cohort, PCGA-Net achieved 28.42 ± 4.88 HU MAE, 0.91 ± 0.05 SSIM, and 38.52 ± 0.95 dB PSNR. In all three cohorts, PCGA-Net exceeded cGAN and PatRecon-Net across all evaluated metrics. 
Significance: These results indicate that jointly enforcing projection-domain completion, analytic imaging geometry, and image-domain refinement is an effective strategy for limited-angle CBCT reconstruction. PCGA-Net provides a robust short-arc reconstruction framework across multiple anatomical sites, supporting its potential utility in constrained imageguided therapeutic and procedural settings.