Deep learning at the edge enables real-time streaming ptychographic imaging

Anakha V. Babu, Tao Zhou, Saugat Kandel, Tekin Bicer, Zhengchun Liu, William Judge, Daniel J. Ching, Yi Jiang, Sinisa Veseli, Steven Henke, Ryan Chard, Yudong Yao, Ekaterina Sirazitdinova, Geetika Gupta, Martin V. Holt, Ian T. Foster, Antonino Miceli () and Mathew J. Cherukara ()
Additional contact information
Anakha V. Babu: Argonne National Laboratory
Tao Zhou: Argonne National Laboratory
Saugat Kandel: Argonne National Laboratory
Tekin Bicer: Argonne National Laboratory
Zhengchun Liu: Argonne National Laboratory
William Judge: University of Illinois
Daniel J. Ching: Argonne National Laboratory
Yi Jiang: Argonne National Laboratory
Sinisa Veseli: Argonne National Laboratory
Steven Henke: Argonne National Laboratory
Ryan Chard: Argonne National Laboratory
Yudong Yao: Argonne National Laboratory
Ekaterina Sirazitdinova: NVIDIA Corporation
Geetika Gupta: NVIDIA Corporation
Martin V. Holt: Argonne National Laboratory
Ian T. Foster: Argonne National Laboratory
Antonino Miceli: Argonne National Laboratory
Mathew J. Cherukara: Argonne National Laboratory

Nature Communications, 2023, vol. 14, issue 1, 1-9

Abstract: Abstract Coherent imaging techniques provide an unparalleled multi-scale view of materials across scientific and technological fields, from structural materials to quantum devices, from integrated circuits to biological cells. Driven by the construction of brighter sources and high-rate detectors, coherent imaging methods like ptychography are poised to revolutionize nanoscale materials characterization. However, these advancements are accompanied by significant increase in data and compute needs, which precludes real-time imaging, feedback and decision-making capabilities with conventional approaches. Here, we demonstrate a workflow that leverages artificial intelligence at the edge and high-performance computing to enable real-time inversion on X-ray ptychography data streamed directly from a detector at up to 2 kHz. The proposed AI-enabled workflow eliminates the oversampling constraints, allowing low-dose imaging using orders of magnitude less data than required by traditional methods.

Date: 2023
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.nature.com/articles/s41467-023-41496-z Abstract (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:nat:natcom:v:14:y:2023:i:1:d:10.1038_s41467-023-41496-z

Ordering information: This journal article can be ordered from
https://www.nature.com/ncomms/

DOI: 10.1038/s41467-023-41496-z

Access Statistics for this article

Nature Communications is currently edited by Nathalie Le Bot, Enda Bergin and Fiona Gillespie

More articles in Nature Communications from Nature
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().