Deep neural network ensembles for THz-TDS refractive index extraction exhibiting resilience to experimental and analytical errors

Terahertz time-domain spectroscopy (THz-TDS) achieves excellent signal-to-noise ratios by measuring the amplitude of the electric field in the time-domain, resulting in the full, complex, frequency-domain information of materials' optical parameters, such as the refractive index. However the da...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Optics express 2023-12, Vol.31 (26), p.44575-44587
Hauptverfasser: Klokkou, Nicholas, Gorecki, Jon, Beddoes, Ben, Apostolopoulos, Vasilis
Format: Artikel
Sprache:eng
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page 44587
container_issue 26
container_start_page 44575
container_title Optics express
container_volume 31
creator Klokkou, Nicholas
Gorecki, Jon
Beddoes, Ben
Apostolopoulos, Vasilis
description Terahertz time-domain spectroscopy (THz-TDS) achieves excellent signal-to-noise ratios by measuring the amplitude of the electric field in the time-domain, resulting in the full, complex, frequency-domain information of materials' optical parameters, such as the refractive index. However the data extraction process is non-trivial and standardization of practices are still yet to be cemented in the field leading to significant variation in sample measurements. One such contribution is low frequency noise offsetting the phase reconstruction of the Fourier transformed signal. Additionally, experimental errors such as fluctuations in the power of the laser driving the spectrometer (laser drift) can heavily contribute to erroneous measurements if not accounted for. We show that ensembles of deep neural networks trained with synthetic data extract the frequency-dependent complex refractive index, whereby required fitting steps are automated and show resilience to phase unwrapping variations and laser drift. We show that training with synthetic data allows for flexibility in the functionality of networks yet the produced ensemble supersedes current extraction techniques.
doi_str_mv 10.1364/OE.507439
format Article
fullrecord <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_miscellaneous_2928957079</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2928957079</sourcerecordid><originalsourceid>FETCH-LOGICAL-c280t-fec84b3490b8a8e7e259a93a1e75ebb09eae4f597ca5d42ec58359cdf02423693</originalsourceid><addsrcrecordid>eNpNkM1OAjEUhRujEUQXvoCZpS4GO9OWtksDKCYkLMT1pNO5o9Whg21RcOebWwSNi5v79-Uk5yB0nuF-Rgb0ejbuM8wpkQeom2FJU4oFP_w3d9CJ9y8YZ5RLfow6RGRcsJx10dcIYJlYWDnVxBY-WveagPWwKBvwSd26ZD75TOejh8RB7ZQO5h0SYytYJ7AOP4fWxvHZlCYY-xQxbxoDVkMS2vhYgjMLsCHqK1vFUs0mGB1XcK51_hQd1arxcLbvPfR4O54PJ-l0dnc_vJmmOhc4pDVoQUtCJS6FEsAhZ1JJojLgDMoSS1BAaya5VqyiOWgmCJO6qnFOczKQpIcud7pL176twIdiYbyGplEW2pUvcpkLyTjmW_Rqh2rXeh99F8voQblNkeFim3gxGxe7xCN7sZddlQuo_sjfiMk3glR-Qg</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2928957079</pqid></control><display><type>article</type><title>Deep neural network ensembles for THz-TDS refractive index extraction exhibiting resilience to experimental and analytical errors</title><source>DOAJ Directory of Open Access Journals</source><source>EZB-FREE-00999 freely available EZB journals</source><source>Alma/SFX Local Collection</source><creator>Klokkou, Nicholas ; Gorecki, Jon ; Beddoes, Ben ; Apostolopoulos, Vasilis</creator><creatorcontrib>Klokkou, Nicholas ; Gorecki, Jon ; Beddoes, Ben ; Apostolopoulos, Vasilis</creatorcontrib><description>Terahertz time-domain spectroscopy (THz-TDS) achieves excellent signal-to-noise ratios by measuring the amplitude of the electric field in the time-domain, resulting in the full, complex, frequency-domain information of materials' optical parameters, such as the refractive index. However the data extraction process is non-trivial and standardization of practices are still yet to be cemented in the field leading to significant variation in sample measurements. One such contribution is low frequency noise offsetting the phase reconstruction of the Fourier transformed signal. Additionally, experimental errors such as fluctuations in the power of the laser driving the spectrometer (laser drift) can heavily contribute to erroneous measurements if not accounted for. We show that ensembles of deep neural networks trained with synthetic data extract the frequency-dependent complex refractive index, whereby required fitting steps are automated and show resilience to phase unwrapping variations and laser drift. We show that training with synthetic data allows for flexibility in the functionality of networks yet the produced ensemble supersedes current extraction techniques.</description><identifier>ISSN: 1094-4087</identifier><identifier>EISSN: 1094-4087</identifier><identifier>DOI: 10.1364/OE.507439</identifier><identifier>PMID: 38178525</identifier><language>eng</language><publisher>United States</publisher><ispartof>Optics express, 2023-12, Vol.31 (26), p.44575-44587</ispartof><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><cites>FETCH-LOGICAL-c280t-fec84b3490b8a8e7e259a93a1e75ebb09eae4f597ca5d42ec58359cdf02423693</cites><orcidid>0000-0002-0999-3745 ; 0000-0003-3733-2191 ; 0000-0001-9205-2294 ; 0000-0003-1577-5362</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>314,780,784,864,27922,27923</link.rule.ids><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/38178525$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Klokkou, Nicholas</creatorcontrib><creatorcontrib>Gorecki, Jon</creatorcontrib><creatorcontrib>Beddoes, Ben</creatorcontrib><creatorcontrib>Apostolopoulos, Vasilis</creatorcontrib><title>Deep neural network ensembles for THz-TDS refractive index extraction exhibiting resilience to experimental and analytical errors</title><title>Optics express</title><addtitle>Opt Express</addtitle><description>Terahertz time-domain spectroscopy (THz-TDS) achieves excellent signal-to-noise ratios by measuring the amplitude of the electric field in the time-domain, resulting in the full, complex, frequency-domain information of materials' optical parameters, such as the refractive index. However the data extraction process is non-trivial and standardization of practices are still yet to be cemented in the field leading to significant variation in sample measurements. One such contribution is low frequency noise offsetting the phase reconstruction of the Fourier transformed signal. Additionally, experimental errors such as fluctuations in the power of the laser driving the spectrometer (laser drift) can heavily contribute to erroneous measurements if not accounted for. We show that ensembles of deep neural networks trained with synthetic data extract the frequency-dependent complex refractive index, whereby required fitting steps are automated and show resilience to phase unwrapping variations and laser drift. We show that training with synthetic data allows for flexibility in the functionality of networks yet the produced ensemble supersedes current extraction techniques.</description><issn>1094-4087</issn><issn>1094-4087</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><recordid>eNpNkM1OAjEUhRujEUQXvoCZpS4GO9OWtksDKCYkLMT1pNO5o9Whg21RcOebWwSNi5v79-Uk5yB0nuF-Rgb0ejbuM8wpkQeom2FJU4oFP_w3d9CJ9y8YZ5RLfow6RGRcsJx10dcIYJlYWDnVxBY-WveagPWwKBvwSd26ZD75TOejh8RB7ZQO5h0SYytYJ7AOP4fWxvHZlCYY-xQxbxoDVkMS2vhYgjMLsCHqK1vFUs0mGB1XcK51_hQd1arxcLbvPfR4O54PJ-l0dnc_vJmmOhc4pDVoQUtCJS6FEsAhZ1JJojLgDMoSS1BAaya5VqyiOWgmCJO6qnFOczKQpIcud7pL176twIdiYbyGplEW2pUvcpkLyTjmW_Rqh2rXeh99F8voQblNkeFim3gxGxe7xCN7sZddlQuo_sjfiMk3glR-Qg</recordid><startdate>20231218</startdate><enddate>20231218</enddate><creator>Klokkou, Nicholas</creator><creator>Gorecki, Jon</creator><creator>Beddoes, Ben</creator><creator>Apostolopoulos, Vasilis</creator><scope>NPM</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7X8</scope><orcidid>https://orcid.org/0000-0002-0999-3745</orcidid><orcidid>https://orcid.org/0000-0003-3733-2191</orcidid><orcidid>https://orcid.org/0000-0001-9205-2294</orcidid><orcidid>https://orcid.org/0000-0003-1577-5362</orcidid></search><sort><creationdate>20231218</creationdate><title>Deep neural network ensembles for THz-TDS refractive index extraction exhibiting resilience to experimental and analytical errors</title><author>Klokkou, Nicholas ; Gorecki, Jon ; Beddoes, Ben ; Apostolopoulos, Vasilis</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c280t-fec84b3490b8a8e7e259a93a1e75ebb09eae4f597ca5d42ec58359cdf02423693</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Klokkou, Nicholas</creatorcontrib><creatorcontrib>Gorecki, Jon</creatorcontrib><creatorcontrib>Beddoes, Ben</creatorcontrib><creatorcontrib>Apostolopoulos, Vasilis</creatorcontrib><collection>PubMed</collection><collection>CrossRef</collection><collection>MEDLINE - Academic</collection><jtitle>Optics express</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Klokkou, Nicholas</au><au>Gorecki, Jon</au><au>Beddoes, Ben</au><au>Apostolopoulos, Vasilis</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Deep neural network ensembles for THz-TDS refractive index extraction exhibiting resilience to experimental and analytical errors</atitle><jtitle>Optics express</jtitle><addtitle>Opt Express</addtitle><date>2023-12-18</date><risdate>2023</risdate><volume>31</volume><issue>26</issue><spage>44575</spage><epage>44587</epage><pages>44575-44587</pages><issn>1094-4087</issn><eissn>1094-4087</eissn><abstract>Terahertz time-domain spectroscopy (THz-TDS) achieves excellent signal-to-noise ratios by measuring the amplitude of the electric field in the time-domain, resulting in the full, complex, frequency-domain information of materials' optical parameters, such as the refractive index. However the data extraction process is non-trivial and standardization of practices are still yet to be cemented in the field leading to significant variation in sample measurements. One such contribution is low frequency noise offsetting the phase reconstruction of the Fourier transformed signal. Additionally, experimental errors such as fluctuations in the power of the laser driving the spectrometer (laser drift) can heavily contribute to erroneous measurements if not accounted for. We show that ensembles of deep neural networks trained with synthetic data extract the frequency-dependent complex refractive index, whereby required fitting steps are automated and show resilience to phase unwrapping variations and laser drift. We show that training with synthetic data allows for flexibility in the functionality of networks yet the produced ensemble supersedes current extraction techniques.</abstract><cop>United States</cop><pmid>38178525</pmid><doi>10.1364/OE.507439</doi><tpages>13</tpages><orcidid>https://orcid.org/0000-0002-0999-3745</orcidid><orcidid>https://orcid.org/0000-0003-3733-2191</orcidid><orcidid>https://orcid.org/0000-0001-9205-2294</orcidid><orcidid>https://orcid.org/0000-0003-1577-5362</orcidid><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier ISSN: 1094-4087
ispartof Optics express, 2023-12, Vol.31 (26), p.44575-44587
issn 1094-4087
1094-4087
language eng
recordid cdi_proquest_miscellaneous_2928957079
source DOAJ Directory of Open Access Journals; EZB-FREE-00999 freely available EZB journals; Alma/SFX Local Collection
title Deep neural network ensembles for THz-TDS refractive index extraction exhibiting resilience to experimental and analytical errors
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-09T11%3A46%3A33IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Deep%20neural%20network%20ensembles%20for%20THz-TDS%20refractive%20index%20extraction%20exhibiting%20resilience%20to%20experimental%20and%20analytical%20errors&rft.jtitle=Optics%20express&rft.au=Klokkou,%20Nicholas&rft.date=2023-12-18&rft.volume=31&rft.issue=26&rft.spage=44575&rft.epage=44587&rft.pages=44575-44587&rft.issn=1094-4087&rft.eissn=1094-4087&rft_id=info:doi/10.1364/OE.507439&rft_dat=%3Cproquest_cross%3E2928957079%3C/proquest_cross%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2928957079&rft_id=info:pmid/38178525&rfr_iscdi=true