-
Physics-assisted Generative Adversarial Network for X-Ray Tomography
Authors:
Zhen Guo,
Jung Ki Song,
George Barbastathis,
Michael E. Glinsky,
Courtenay T. Vaughan,
Kurt W. Larson,
Bradley K. Alpert,
Zachary H. Levine
Abstract:
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.…
▽ More
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.
△ Less
Submitted 3 June, 2022; v1 submitted 7 April, 2022;
originally announced April 2022.
-
Advantage of Machine Learning over Maximum Likelihood in Limited-Angle Low-Photon X-Ray Tomography
Authors:
Zhen Guo,
Jung Ki Song,
George Barbastathis,
Michael E. Glinsky,
Courtenay T. Vaughan,
Kurt W. Larson,
Bradley K. Alpert,
Zachary H. Levine
Abstract:
Limited-angle X-ray tomography reconstruction is an ill-conditioned inverse problem in general. Especially when the projection angles are limited and the measurements are taken in a photon-limited condition, reconstructions from classical algorithms such as filtered backprojection may lose fidelity and acquire artifacts due to the missing-cone problem. To obtain satisfactory reconstruction results…
▽ More
Limited-angle X-ray tomography reconstruction is an ill-conditioned inverse problem in general. Especially when the projection angles are limited and the measurements are taken in a photon-limited condition, reconstructions from classical algorithms such as filtered backprojection may lose fidelity and acquire artifacts due to the missing-cone problem. To obtain satisfactory reconstruction results, prior assumptions, such as total variation minimization and nonlocal image similarity, are usually incorporated within the reconstruction algorithm. In this work, we introduce deep neural networks to determine and apply a prior distribution in the reconstruction process. Our neural networks learn the prior directly from synthetic training samples. The neural nets thus obtain a prior distribution that is specific to the class of objects we are interested in reconstructing. In particular, we used deep generative models with 3D convolutional layers and 3D attention layers which are trained on 3D synthetic integrated circuit (IC) data from a model dubbed CircuitFaker. We demonstrate that, when the projection angles and photon budgets are limited, the priors from our deep generative models can dramatically improve the IC reconstruction quality on synthetic data compared with maximum likelihood estimation. Training the deep generative models with synthetic IC data from CircuitFaker illustrates the capabilities of the learned prior from machine learning. We expect that if the process were reproduced with experimental data, the advantage of the machine learning would persist. The advantages of machine learning in limited angle X-ray tomography may further enable applications in low-photon nanoscale imaging.
△ Less
Submitted 18 December, 2021; v1 submitted 15 November, 2021;
originally announced November 2021.
-
Towards an Open and Scalable Music Metadata Layer
Authors:
Thomas Hardjono,
George Howard,
Eric Scace,
Mizan Chowdury,
Lucas Novak,
Meghan Gaudet,
Justin Anderson,
Nicole d'avis,
Christopher Kulis,
Edward Sweeney,
Chandler Vaughan
Abstract:
One of the significant issues in the music supply chain today is the lack of consistent, complete and authoritative information or metadata regarding the creation of a given musical work. In many cases multiple entities in the music supply chain have each created their own version of the metadata for a musical work, often by manually re-entering the same information or through scra** data from o…
▽ More
One of the significant issues in the music supply chain today is the lack of consistent, complete and authoritative information or metadata regarding the creation of a given musical work. In many cases multiple entities in the music supply chain have each created their own version of the metadata for a musical work, often by manually re-entering the same information or through scra** data from other sites. In such cases, the effort to synchronize or to correct the information becomes manually laborious and error-prone. Furthermore, confidential information regarding the legal ownership of the musical work is often commingled in the same metadata, making the entire database proprietary and thus closed. In this paper we explore an alternative model for creation metadata following the open access paradigm found in other industries, such as in book publishing, library systems and in the automotive parts supply chain. The vision is to create a new music metadata layer for creation metadata that is open, scalable and provides an authoritative source of information that is available to all entities in the music supply chain globally.
△ Less
Submitted 7 November, 2019;
originally announced November 2019.
