Physics-assisted Generative Adversarial Network for X-Ray Tomography
This addresses the challenge of low-photon imaging in fields like biomedical imaging and materials science, offering an incremental improvement over existing deep learning methods.
The paper tackles the ill-conditioned inverse problem in X-ray tomography reconstruction by developing a Physics-assisted Generative Adversarial Network (PGAN) that uses maximum-likelihood estimates to regularize with known physics and a learned prior, reducing photon requirements with limited projection angles to achieve a given error rate.
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.