Deep Learning Accelerated Light Source Experiments

Liu, Zhengchun; Biçer, Tekin; Kettimuthu, Rajkumar; Foster, Ian

doi:10.1109/dls49591.2019.00008

Cited by 29 publications

(16 citation statements)

References 52 publications

(61 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…SSIM is improved significantly by TomoGAN and is improved consistently across all the slices, suggesting that the method is reliable. The evaluation of this low-dose scenario can also be used for streaming tomography where the views are very limited at the beginning of experimentation (e.g., Liu et al 2019 [55] 128 projections 64 projections…”

Section: Fewer Projectionsmentioning

confidence: 99%

TomoGAN: low-dose synchrotron x-ray tomography with generative adversarial networks: discussion

et al. 2020

Self Cite

View full text Add to dashboard Cite

Synchrotron-based x-ray tomography is a noninvasive imaging technique that allows for reconstructing the internal structure of materials at high spatial resolutions from tens of micrometers to a few nanometers. In order to resolve sample features at smaller length scales, however, a higher radiation dose is required. Therefore, the limitation on the achievable resolution is set primarily by noise at these length scales. We present TomoGAN, a denoising technique based on generative adversarial networks, for improving the quality of reconstructed images for low-dose imaging conditions. We evaluate our approach in two photon-budget-limited experimental conditions: (1) sufficient number of low-dose projections (based on Nyquist sampling), and (2) insufficient or limited number of high-dose projections. In both cases the angular sampling is assumed to be isotropic, and the photon budget throughout the experiment is fixed based on the maximum allowable radiation dose on the sample. Evaluation with both simulated and experimental datasets shows that our approach can significantly reduce noise in reconstructed images, improving the structural similarity score of simulation and experimental data from 0.18 to 0.9 and from 0.18 to 0.41, respectively. Furthermore, the quality of the reconstructed images with filtered back projection followed by our denoising approach exceeds that of reconstructions with the simultaneous iterative reconstruction technique, showing the computational superiority of our approach.

show abstract

Section: Fewer Projectionsmentioning

confidence: 99%

TomoGAN: low-dose synchrotron x-ray tomography with generative adversarial networks: discussion

et al. 2020

Self Cite

View full text Add to dashboard Cite

show abstract

“…To explore whether a particular deep learning model can provide sufficient accuracy on edge devices, TomoGAN (94,95), which is an algorithm for enhancing the quality of X-ray images, was adapted to run on the Google Edge TPU and NVIDIA Jetson TX2 (96). The benchmarking results indicated that edge accelerators can provide sufficient accuracy with a novel shallow CNN called the fine-tune network.…”

Section: System Under Testmentioning

confidence: 99%

A Survey on Edge Performance Benchmarking

et al. 2021

View full text Add to dashboard Cite

Edge computing is the next Internet frontier that will leverage computing resources located near users, sensors, and data stores to provide more responsive services. Therefore, it is envisioned that a large-scale, geographically dispersed, and resource-rich distributed system will emerge and play a key role in the future Internet. However, given the loosely coupled nature of such complex systems, their operational conditions are expected to change significantly over time. In this context, the performance characteristics of such systems will need to be captured rapidly, which is referred to as performance benchmarking, for application deployment, resource orchestration, and adaptive decision-making. Edge performance benchmarking is a nascent research avenue that has started gaining momentum over the past five years. This article first reviews articles published over the past three decades to trace the history of performance benchmarking from tightly coupled to loosely coupled systems. It then systematically classifies previous research to identify the system under test, techniques analyzed, and benchmark runtime in edge performance benchmarking.

show abstract

“…Many machine learning (ML) techniques have been successfully used and integrated to light source and electron microscopy data analysis workflows to enhance and improve the quality of images and reconstructions [9,36,56], including image denoising [62,39], artifact reduction [65] and feature extraction [51]. These techniques can also be used for accelerating the performance of workflows and data acquisition [38]. We plan to incorporate some of these advanced ML techniques in our workflow in the future.…”

Section: Related Workmentioning

confidence: 99%

High-Performance Ptychographic Reconstruction with Federated Facilities

Biçer¹,

Yu²,

Ching³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Beamlines at synchrotron light source facilities are powerful scientific instruments used to image samples and observe phenomena at high spatial and temporal resolutions. Typically, these facilities are equipped only with modest compute resources for the analysis of generated experimental datasets. However, high data rate experiments can easily generate data in volumes that take days (or even weeks) to process on those local resources. To address this challenge, we present a system that unifies leadership computing and experimental facilities by enabling the automated establishment of data analysis pipelines that extend from edge data acquisition systems at synchrotron beamlines to remote computing facilities; under the covers, our system uses Globus Auth authentication to minimize user interaction, funcX to run user-defined functions on supercomputers, and Globus Flows to define and execute workflows. We describe the application of this system to ptychography, an ultrahigh-resolution coherent diffraction imaging technique that can produce 100s of gigabytes to terabytes in a single experiment. When deployed on the DGX A100 ThetaGPU cluster at the Argonne Leadership Computing Facility and a microscopy beamline at the Advanced Photon Source, our system performs analysis as an experiment progresses to provide timely feedback.

show abstract

Deep Learning Accelerated Light Source Experiments

Cited by 29 publications

References 52 publications

TomoGAN: low-dose synchrotron x-ray tomography with generative adversarial networks: discussion

TomoGAN: low-dose synchrotron x-ray tomography with generative adversarial networks: discussion

A Survey on Edge Performance Benchmarking

High-Performance Ptychographic Reconstruction with Federated Facilities

Contact Info

Product

Resources

About