scholarly journals Automated analysis of hot spot X-ray images at the National Ignition Facility

2016 ◽  
Vol 87 (11) ◽  
pp. 11E334 ◽  
Author(s):  
S. F. Khan ◽  
N. Izumi ◽  
S. Glenn ◽  
R. Tommasini ◽  
L. R. Benedetti ◽  
...  
2018 ◽  
Vol 89 (10) ◽  
pp. 10G111 ◽  
Author(s):  
D. T. Bishel ◽  
B. Bachmann ◽  
A. Yi ◽  
D. Kraus ◽  
L. Divol ◽  
...  

2021 ◽  
Vol 103 (2) ◽  
Author(s):  
R. C. Shah ◽  
S. X. Hu ◽  
I. V. Igumenshchev ◽  
J. Baltazar ◽  
D. Cao ◽  
...  

2019 ◽  
Vol 26 (6) ◽  
pp. 063105 ◽  
Author(s):  
M. J. May ◽  
G. E. Kemp ◽  
J. D. Colvin ◽  
D. A. Liedahl ◽  
P. L. Poole ◽  
...  

2010 ◽  
Vol 17 (8) ◽  
pp. 082701 ◽  
Author(s):  
K. B. Fournier ◽  
M. J. May ◽  
J. D. Colvin ◽  
J. O. Kane ◽  
M. Schneider ◽  
...  

1999 ◽  
Vol 17 (2) ◽  
pp. 217-224 ◽  
Author(s):  
T.R. DITTRICH ◽  
S.W. HAAN ◽  
M.M. MARINAK ◽  
D.E. HINKEL ◽  
S.M. POLLAINE ◽  
...  

Several choices exist in the design and production of capsules intended to ignite and propagate fusion burn of the deuterium–tritium (D–T) fuel when imploded by indirect drive at the National Ignition Facility (NIF). These choices include ablator material, ablator dopant concentration and distribution, capsule dimensions, and X-ray drive profile (shock timings and strengths). The choice of ablator material must also include fabrication and material characteristics, such as attainable surface finishes, permeability, strength, transparency to radio frequency and infrared radiation, thermal conductivity, and material homogeneity. Understanding the advantages and/or limitations of these choices is an ongoing effort for LLNL and LANL designers. At this time, simulations in one-, two-, and three-dimensions show that capsules with either a copper-doped beryllium or a polyimide (C22H10N2O4) ablator material have both the least sensitivity to initial surface roughnesses and favorable fabrication qualities. Simulations also indicate the existence of capsule designs based on these ablator materials which ignite and burn when imploded by less than nominal laser performance (900-kJ energy, 250-TW power, producing 250-eV peak radiation temperature). We will describe and compare these reduced-scale capsules, in addition to several designs which use the expected 300-eV peak X-ray drive obtained from operating the NIF laser at 1.3 MJ and 500 TW.


2018 ◽  
Vol 89 (10) ◽  
pp. 10G121 ◽  
Author(s):  
C. M. Huntington ◽  
J. M. McNaney ◽  
E. Gumbrell ◽  
A. Krygier ◽  
C. Wehrenberg ◽  
...  

2003 ◽  
Vol 9 (1) ◽  
pp. 1-17 ◽  
Author(s):  
Paul G. Kotula ◽  
Michael R. Keenan ◽  
Joseph R. Michael

Spectral imaging in the scanning electron microscope (SEM) equipped with an energy-dispersive X-ray (EDX) analyzer has the potential to be a powerful tool for chemical phase identification, but the large data sets have, in the past, proved too large to efficiently analyze. In the present work, we describe the application of a new automated, unbiased, multivariate statistical analysis technique to very large X-ray spectral image data sets. The method, based in part on principal components analysis, returns physically accurate (all positive) component spectra and images in a few minutes on a standard personal computer. The efficacy of the technique for microanalysis is illustrated by the analysis of complex multi-phase materials, particulates, a diffusion couple, and a single-pixel-detection problem.


2021 ◽  
Author(s):  
Md Inzamam Ul Haque ◽  
Abhishek K Dubey ◽  
Jacob D Hinkle

Deep learning models have received much attention lately for their ability to achieve expert-level performance on the accurate automated analysis of chest X-rays. Although publicly available chest X-ray datasets include high resolution images, most models are trained on reduced size images due to limitations on GPU memory and training time. As compute capability continues to advance, it will become feasible to train large convolutional neural networks on high-resolution images. This study is based on the publicly available MIMIC-CXR-JPG dataset, comprising 377,110 high resolution chest X-ray images, and provided with 14 labels to the corresponding free-text radiology reports. We find, interestingly, that tasks that require a large receptive field are better suited to downscaled input images, and we verify this qualitatively by inspecting effective receptive fields and class activation maps of trained models. Finally, we show that stacking an ensemble across resolutions outperforms each individual learner at all input resolutions while providing interpretable scale weights, suggesting that multi-scale features are crucially important to information extraction from high-resolution chest X-rays.


Sign in / Sign up

Export Citation Format

Share Document