Physics-informed Machine Learning for High-resolution X-Ray Imaging

The recent inertial confinement fusion (ICF) breakthroughs motivate high-resolution X-ray imaging [1]. In addition, X-ray imaging is commonly used for material identification through X-ray fluorescence, dose control in proton therapy, and computed tomography (CT) scans. We use physics-informed synthetic datasets to train a convolution neural network (CNN) to achieve high X-ray imaging resolution. The physics-informed synthetic data sets are generated using the Allpix squared simulation framework, extending our recent work that used a fully connected neural network (FCNN) [2-3]. The physics-informed CNN algorithm is further validated using experimental X-ray datasets from multiple X-ray sources as described in [4]. Applications of the physics-informed CNN algorithm to proton therapy experiments will also be presented. Our work also paves way towards high-resolution X-ray imaging for ICF, magnetic fusion (impurity control) and commercial CT applications.

[1] H Abu-Shawareb et al., ‘Lawson criterion for ignition exceeded in an inertial fusion experiment,’ Physical Review Letters vol. 129, 075001 (2022).

[2] S. Lin, J. K. Baldwin, M. Blatnik, S. M. Clayton, C. Cude-Woods, S. A. Currie, B. Filippone, E. M. Fries, P. Geltenbort, A. T. Holley, W. Li, C.-Y. Liu, M. Makela, C. L. Morris, R. Musedinovic, C. O’Shaughnessy, R. W. Pattie Jr., D. J. Salvat, A. Saunders, E. I. Sharapov, M. Singh, X. Sun, Z. Tang, W. Uhrich, W. Wei, B. Wolfe, A.R. Young, Z. Wang, ‘Demonstration of Sub-micron UCN Position Resolution using Room-temperature CMOS Sensor,’ submitted; preprint at arXiv:2305.09562 (2023);

[3] X. Yue, S. Lin, W. Li et al., ‘Ultrafast CMOS image sensors and data-enabled super-resolution for multimodal radiographic imaging and tomography.’ PoS vol. 420, 0041 (2023).

[4] Z. Wang, K. Anagnost, C. W. Barnes, D. M. Dattelbaum, E. R. Fossum, E. Lee, J. Liu, J. J. Ma, W. Z. Meijer, W. Nie, C. M. Sweeney, A. C. Therrien, H. Tsai, X. Yue, ‘Billion-pixel x-ray camera (BiPC-X)’, Rev Sci Instrum 92 (4), 043708, (2021). doi.org/10.1063/5.0043013.