X-ray tomography is capable of imaging the interior of objects in three dimensions non-invasively, with applications in biomedical imaging, materials science, electronic inspection, and other fields. The reconstruction process can be an ill-conditioned inverse problem, requiring regularization to obtain satisfactory results. Recently, deep learning has been adopted for tomographic reconstruction. Unlike iterative algorithms which require a distribution that is known a priori , deep reconstruction networks can learn a prior distribution through sampling the training distributions. In this work, we develop a Physics-assisted Generative Adversarial Network (PGAN), a two-step algorithm for tomographic reconstruction. In contrast to previous efforts, our PGAN utilizes maximum-likelihood estimates derived from the measurements to regularize the reconstruction with both known physics and the learned prior. Compared with methods with less physics assisting in training, PGAN can reduce the photon requirement with limited projection angles to achieve a given error rate. The advantages of using a physics-assisted learned prior in X-ray tomography may further enable low-photon nanoscale imaging.
Guo, Zhen, Song, Jung Ki, Barbastathis, George, Glinsky, Michael E., Vaughan, Courtenay T., Larson, Kurt W., Alpert, Bradley K., & Levine, Zachary H. (2022). Physics-assisted generative adversarial network for X-ray tomography. Optics Express, 30(13). https://doi.org/10.1364/OE.460208
@article{osti_1871842,
author = {Guo, Zhen and Song, Jung Ki and Barbastathis, George and Glinsky, Michael E. and Vaughan, Courtenay T. and Larson, Kurt W. and Alpert, Bradley K. and Levine, Zachary H.},
title = {Physics-assisted generative adversarial network for X-ray tomography},
annote = { X-ray tomography is capable of imaging the interior of objects in three dimensions non-invasively, with applications in biomedical imaging, materials science, electronic inspection, and other fields. The reconstruction process can be an ill-conditioned inverse problem, requiring regularization to obtain satisfactory results. Recently, deep learning has been adopted for tomographic reconstruction. Unlike iterative algorithms which require a distribution that is known a priori , deep reconstruction networks can learn a prior distribution through sampling the training distributions. In this work, we develop a Physics-assisted Generative Adversarial Network (PGAN), a two-step algorithm for tomographic reconstruction. In contrast to previous efforts, our PGAN utilizes maximum-likelihood estimates derived from the measurements to regularize the reconstruction with both known physics and the learned prior. Compared with methods with less physics assisting in training, PGAN can reduce the photon requirement with limited projection angles to achieve a given error rate. The advantages of using a physics-assisted learned prior in X-ray tomography may further enable low-photon nanoscale imaging. },
doi = {10.1364/OE.460208},
url = {https://www.osti.gov/biblio/1871842},
journal = {Optics Express},
issn = {ISSN OPEXFF},
number = {13},
volume = {30},
place = {United States},
publisher = {Optical Society of America},
year = {2022},
month = {06}}
Kazantsev, D.; Van Eyndhoven, G.; Lionheart, W. R. B.
Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, Vol. 373, Issue 2043https://doi.org/10.1098/rsta.2014.0389