Low-dose x-ray tomography through a deep convolutional neural network
Abstract Synchrotron-based X-ray tomography offers the potential for rapid large-scale reconstructions of the interiors of materials and biological tissue at fine resolution. However, for radiation sensitive samples, there remain fundamental trade-offs between damaging samples during longer acquisit...
Guardado en:
Autores principales: | , , , , , , |
---|---|
Formato: | article |
Lenguaje: | EN |
Publicado: |
Nature Portfolio
2018
|
Materias: | |
Acceso en línea: | https://doaj.org/article/f2c3eb2baba2489aa424e39aaef3f1eb |
Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
id |
oai:doaj.org-article:f2c3eb2baba2489aa424e39aaef3f1eb |
---|---|
record_format |
dspace |
spelling |
oai:doaj.org-article:f2c3eb2baba2489aa424e39aaef3f1eb2021-12-02T15:08:02ZLow-dose x-ray tomography through a deep convolutional neural network10.1038/s41598-018-19426-72045-2322https://doaj.org/article/f2c3eb2baba2489aa424e39aaef3f1eb2018-02-01T00:00:00Zhttps://doi.org/10.1038/s41598-018-19426-7https://doaj.org/toc/2045-2322Abstract Synchrotron-based X-ray tomography offers the potential for rapid large-scale reconstructions of the interiors of materials and biological tissue at fine resolution. However, for radiation sensitive samples, there remain fundamental trade-offs between damaging samples during longer acquisition times and reducing signals with shorter acquisition times. We present a deep convolutional neural network (CNN) method that increases the acquired X-ray tomographic signal by at least a factor of 10 during low-dose fast acquisition by improving the quality of recorded projections. Short-exposure-time projections enhanced with CNNs show signal-to-noise ratios similar to long-exposure-time projections. They also show lower noise and more structural information than low-dose short-exposure acquisitions post-processed by other techniques. We evaluated this approach using simulated samples and further validated it with experimental data from radiation sensitive mouse brains acquired in a tomographic setting with transmission X-ray microscopy. We demonstrate that automated algorithms can reliably trace brain structures in low-dose datasets enhanced with CNN. This method can be applied to other tomographic or scanning based X-ray imaging techniques and has great potential for studying faster dynamics in specimensXiaogang YangVincent De AndradeWilliam ScullinEva L. DyerNarayanan KasthuriFrancesco De CarloDoğa GürsoyNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 8, Iss 1, Pp 1-13 (2018) |
institution |
DOAJ |
collection |
DOAJ |
language |
EN |
topic |
Medicine R Science Q |
spellingShingle |
Medicine R Science Q Xiaogang Yang Vincent De Andrade William Scullin Eva L. Dyer Narayanan Kasthuri Francesco De Carlo Doğa Gürsoy Low-dose x-ray tomography through a deep convolutional neural network |
description |
Abstract Synchrotron-based X-ray tomography offers the potential for rapid large-scale reconstructions of the interiors of materials and biological tissue at fine resolution. However, for radiation sensitive samples, there remain fundamental trade-offs between damaging samples during longer acquisition times and reducing signals with shorter acquisition times. We present a deep convolutional neural network (CNN) method that increases the acquired X-ray tomographic signal by at least a factor of 10 during low-dose fast acquisition by improving the quality of recorded projections. Short-exposure-time projections enhanced with CNNs show signal-to-noise ratios similar to long-exposure-time projections. They also show lower noise and more structural information than low-dose short-exposure acquisitions post-processed by other techniques. We evaluated this approach using simulated samples and further validated it with experimental data from radiation sensitive mouse brains acquired in a tomographic setting with transmission X-ray microscopy. We demonstrate that automated algorithms can reliably trace brain structures in low-dose datasets enhanced with CNN. This method can be applied to other tomographic or scanning based X-ray imaging techniques and has great potential for studying faster dynamics in specimens |
format |
article |
author |
Xiaogang Yang Vincent De Andrade William Scullin Eva L. Dyer Narayanan Kasthuri Francesco De Carlo Doğa Gürsoy |
author_facet |
Xiaogang Yang Vincent De Andrade William Scullin Eva L. Dyer Narayanan Kasthuri Francesco De Carlo Doğa Gürsoy |
author_sort |
Xiaogang Yang |
title |
Low-dose x-ray tomography through a deep convolutional neural network |
title_short |
Low-dose x-ray tomography through a deep convolutional neural network |
title_full |
Low-dose x-ray tomography through a deep convolutional neural network |
title_fullStr |
Low-dose x-ray tomography through a deep convolutional neural network |
title_full_unstemmed |
Low-dose x-ray tomography through a deep convolutional neural network |
title_sort |
low-dose x-ray tomography through a deep convolutional neural network |
publisher |
Nature Portfolio |
publishDate |
2018 |
url |
https://doaj.org/article/f2c3eb2baba2489aa424e39aaef3f1eb |
work_keys_str_mv |
AT xiaogangyang lowdosexraytomographythroughadeepconvolutionalneuralnetwork AT vincentdeandrade lowdosexraytomographythroughadeepconvolutionalneuralnetwork AT williamscullin lowdosexraytomographythroughadeepconvolutionalneuralnetwork AT evaldyer lowdosexraytomographythroughadeepconvolutionalneuralnetwork AT narayanankasthuri lowdosexraytomographythroughadeepconvolutionalneuralnetwork AT francescodecarlo lowdosexraytomographythroughadeepconvolutionalneuralnetwork AT dogagursoy lowdosexraytomographythroughadeepconvolutionalneuralnetwork |
_version_ |
1718388291558965248 |