Deep learning-based protoacoustic signal denoising for proton range verification

Proton therapy is a type of radiation therapy that can provide better dose distribution compared to photon therapy by delivering most of the energy at the end of range, which is called the Bragg peak (BP). The protoacoustic technique was developed to determine the BP locations , but it requires a la...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Biomedical physics & engineering express Ročník 9; číslo 4
Hlavní autoři: Wang, Jing, Sohn, James J, Lei, Yang, Nie, Wei, Zhou, Jun, Avery, Stephen, Liu, Tian, Yang, Xiaofeng
Médium: Journal Article
Jazyk:angličtina
Vydáno: England IOP Publishing 12.05.2023
Témata:
ISSN:2057-1976, 2057-1976
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Abstract Proton therapy is a type of radiation therapy that can provide better dose distribution compared to photon therapy by delivering most of the energy at the end of range, which is called the Bragg peak (BP). The protoacoustic technique was developed to determine the BP locations , but it requires a large dose delivery to the tissue to obtain a high number of signal averaging (NSA) to achieve a sufficient signal-to-noise ratio (SNR), which is not suitable for clinical use. A novel deep learning-based technique has been proposed to denoise acoustic signals and reduce BP range uncertainty with much lower doses. Three accelerometers were placed on the distal surface of a cylindrical polyethylene (PE) phantom to collect protoacoustic signals. In total, 512 raw signals were collected at each device. Device-specific stack autoencoder (SAE) denoising models were trained to denoise the noise-containing input signals, which were generated by averaging only 1, 2, 4, 8, 16, or 24 raw signals (low NSA signals), while the clean signals were obtained by averaging 192 raw signals (high NSA). Both supervised and unsupervised training strategies were employed, and the evaluation of the models was based on mean squared error (MSE), SNR, and BP range uncertainty. Overall, the supervised SAEs outperformed the unsupervised SAEs in BP range verification. For the high accuracy detector, it achieved a BP range uncertainty of 0.20 ± 3.44 mm by averaging over 8 raw signals, while for the other two low accuracy detectors, they achieved the BP uncertainty of 1.44 ± 6.45 mm and -0.23 ± 4.88 mm by averaging 16 raw signals, respectively. This deep learning-based denoising method has shown promising results in enhancing the SNR of protoacoustic measurements and improving the accuracy in BP range verification. It greatly reduces the dose and time for potential clinical applications.
AbstractList Proton therapy is a type of radiation therapy that can provide better dose distribution compared to photon therapy by delivering most of the energy at the end of range, which is called the Bragg peak (BP). The protoacoustic technique was developed to determine the BP locationsin vivo, but it requires a large dose delivery to the tissue to obtain a high number of signal averaging (NSA) to achieve a sufficient signal-to-noise ratio (SNR), which is not suitable for clinical use. A novel deep learning-based technique has been proposed to denoise acoustic signals and reduce BP range uncertainty with much lower doses. Three accelerometers were placed on the distal surface of a cylindrical polyethylene (PE) phantom to collect protoacoustic signals. In total, 512 raw signals were collected at each device. Device-specific stack autoencoder (SAE) denoising models were trained to denoise the noise-containing input signals, which were generated by averaging only 1, 2, 4, 8, 16, or 24 raw signals (low NSA signals), while the clean signals were obtained by averaging 192 raw signals (high NSA). Both supervised and unsupervised training strategies were employed, and the evaluation of the models was based on mean squared error (MSE), SNR, and BP range uncertainty. Overall, the supervised SAEs outperformed the unsupervised SAEs in BP range verification. For the high accuracy detector, it achieved a BP range uncertainty of 0.20 ± 3.44 mm by averaging over 8 raw signals, while for the other two low accuracy detectors, they achieved the BP uncertainty of 1.44 ± 6.45 mm and -0.23 ± 4.88 mm by averaging 16 raw signals, respectively. This deep learning-based denoising method has shown promising results in enhancing the SNR of protoacoustic measurements and improving the accuracy in BP range verification. It greatly reduces the dose and time for potential clinical applications.Proton therapy is a type of radiation therapy that can provide better dose distribution compared to photon therapy by delivering most of the energy at the end of range, which is called the Bragg peak (BP). The protoacoustic technique was developed to determine the BP locationsin vivo, but it requires a large dose delivery to the tissue to obtain a high number of signal averaging (NSA) to achieve a sufficient signal-to-noise ratio (SNR), which is not suitable for clinical use. A novel deep learning-based technique has been proposed to denoise acoustic signals and reduce BP range uncertainty with much lower doses. Three accelerometers were placed on the distal surface of a cylindrical polyethylene (PE) phantom to collect protoacoustic signals. In total, 512 raw signals were collected at each device. Device-specific stack autoencoder (SAE) denoising models were trained to denoise the noise-containing input signals, which were generated by averaging only 1, 2, 4, 8, 16, or 24 raw signals (low NSA signals), while the clean signals were obtained by averaging 192 raw signals (high NSA). Both supervised and unsupervised training strategies were employed, and the evaluation of the models was based on mean squared error (MSE), SNR, and BP range uncertainty. Overall, the supervised SAEs outperformed the unsupervised SAEs in BP range verification. For the high accuracy detector, it achieved a BP range uncertainty of 0.20 ± 3.44 mm by averaging over 8 raw signals, while for the other two low accuracy detectors, they achieved the BP uncertainty of 1.44 ± 6.45 mm and -0.23 ± 4.88 mm by averaging 16 raw signals, respectively. This deep learning-based denoising method has shown promising results in enhancing the SNR of protoacoustic measurements and improving the accuracy in BP range verification. It greatly reduces the dose and time for potential clinical applications.
Proton therapy is a type of radiation therapy that can provide better dose distribution compared to photon therapy by delivering most of the energy at the end of range, which is called the Bragg peak (BP). The protoacoustic technique was developed to determine the BP locations , but it requires a large dose delivery to the tissue to obtain a high number of signal averaging (NSA) to achieve a sufficient signal-to-noise ratio (SNR), which is not suitable for clinical use. A novel deep learning-based technique has been proposed to denoise acoustic signals and reduce BP range uncertainty with much lower doses. Three accelerometers were placed on the distal surface of a cylindrical polyethylene (PE) phantom to collect protoacoustic signals. In total, 512 raw signals were collected at each device. Device-specific stack autoencoder (SAE) denoising models were trained to denoise the noise-containing input signals, which were generated by averaging only 1, 2, 4, 8, 16, or 24 raw signals (low NSA signals), while the clean signals were obtained by averaging 192 raw signals (high NSA). Both supervised and unsupervised training strategies were employed, and the evaluation of the models was based on mean squared error (MSE), SNR, and BP range uncertainty. Overall, the supervised SAEs outperformed the unsupervised SAEs in BP range verification. For the high accuracy detector, it achieved a BP range uncertainty of 0.20 ± 3.44 mm by averaging over 8 raw signals, while for the other two low accuracy detectors, they achieved the BP uncertainty of 1.44 ± 6.45 mm and -0.23 ± 4.88 mm by averaging 16 raw signals, respectively. This deep learning-based denoising method has shown promising results in enhancing the SNR of protoacoustic measurements and improving the accuracy in BP range verification. It greatly reduces the dose and time for potential clinical applications.
Author Nie, Wei
Yang, Xiaofeng
Zhou, Jun
Avery, Stephen
Liu, Tian
Sohn, James J
Wang, Jing
Lei, Yang
Author_xml – sequence: 1
  givenname: Jing
  orcidid: 0000-0003-1618-2016
  surname: Wang
  fullname: Wang, Jing
  organization: Emory University Department of Radiation Oncology and Winship Cancer Institute, Atlanta, GA, United States of America
– sequence: 2
  givenname: James J
  surname: Sohn
  fullname: Sohn, James J
  organization: Northwestern University Department of Radiation Oncology, Chicago IL, United States of America
– sequence: 3
  givenname: Yang
  surname: Lei
  fullname: Lei, Yang
  organization: Emory University Department of Radiation Oncology and Winship Cancer Institute, Atlanta, GA, United States of America
– sequence: 4
  givenname: Wei
  surname: Nie
  fullname: Nie, Wei
  organization: Radiation Oncology Division, Inova Schar Cancer Institute, Fairfax, VA, United States of America
– sequence: 5
  givenname: Jun
  surname: Zhou
  fullname: Zhou, Jun
  organization: Emory University Department of Radiation Oncology and Winship Cancer Institute, Atlanta, GA, United States of America
– sequence: 6
  givenname: Stephen
  surname: Avery
  fullname: Avery, Stephen
  organization: University of Pennsylvania Department of Radiation Oncology, Philadelphia, PA, United States of America
– sequence: 7
  givenname: Tian
  surname: Liu
  fullname: Liu, Tian
  organization: Department of Radiation Oncology, Mount Sinai Medical Center, New York, NY 10029, United States of America
– sequence: 8
  givenname: Xiaofeng
  orcidid: 0000-0001-9023-5855
  surname: Yang
  fullname: Yang, Xiaofeng
  organization: Emory University Department of Radiation Oncology and Winship Cancer Institute, Atlanta, GA, United States of America
BackLink https://www.ncbi.nlm.nih.gov/pubmed/37141867$$D View this record in MEDLINE/PubMed
BookMark eNptkb1PwzAQxS0EoqV0Z0LZYCDUdhI7HlH5lCrBALPl2OfKVWqHOEHw35MqBTEw3enpp9O7907QoQ8eEDoj-JrgslxQXPCUCM4WShta8AM0_ZUO_-wTNI9xgzEmjDImimM0yTjJScn4FL3cAjRJDar1zq_TSkUwSdOGLigd-tg5nUS39qpODPjg4gAlNrQj4pNW-TUkH9A667TqXPCn6MiqOsJ8P2fo7f7udfmYrp4fnpY3q1TnjHUp00LnmqhCW1Fyozm1IDhkzFBWCT6IGTWFzgRgmpcWsLGkhAyXAhQRuchm6HK8Ozh57yF2cuuihrpWHgbjkpYEC5rllA3o-R7tqy0Y2bRuq9ov-ZPCAFyMgAuN3IS-Hf6NsmrgUwqZS5wXGDPZGDuQV_-QBMtdI3IXudxFLsdGsm-qCH2Q
CODEN NJOPFM
CitedBy_id crossref_primary_10_1002_mp_17509
crossref_primary_10_1088_2057_1976_adeb90
crossref_primary_10_1002_mp_17079
crossref_primary_10_1002_mp_17600
crossref_primary_10_1088_1361_6560_ad0283
crossref_primary_10_3390_polym17152068
ContentType Journal Article
Copyright 2023 The Author(s). Published by IOP Publishing Ltd
Creative Commons Attribution license.
Copyright_xml – notice: 2023 The Author(s). Published by IOP Publishing Ltd
– notice: Creative Commons Attribution license.
DBID O3W
TSCCA
CGR
CUY
CVF
ECM
EIF
NPM
7X8
DOI 10.1088/2057-1976/acd257
DatabaseName Institute of Physics Open Access Journal Titles
IOPscience (Open Access)
Medline
MEDLINE
MEDLINE (Ovid)
MEDLINE
MEDLINE
PubMed
MEDLINE - Academic
DatabaseTitle MEDLINE
Medline Complete
MEDLINE with Full Text
PubMed
MEDLINE (Ovid)
MEDLINE - Academic
DatabaseTitleList MEDLINE - Academic
MEDLINE
Database_xml – sequence: 1
  dbid: NPM
  name: PubMed
  url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed
  sourceTypes: Index Database
– sequence: 2
  dbid: O3W
  name: Institute of Physics Open Access Journal Titles
  url: http://iopscience.iop.org/
  sourceTypes:
    Enrichment Source
    Publisher
– sequence: 3
  dbid: 7X8
  name: MEDLINE - Academic
  url: https://search.proquest.com/medline
  sourceTypes: Aggregation Database
DeliveryMethod fulltext_linktorsrc
EISSN 2057-1976
ExternalDocumentID 37141867
bpexacd257
Genre Journal Article
Research Support, N.I.H., Extramural
GrantInformation_xml – fundername: National Institutes of Health
  grantid: R01CA215718; R56EB033332
– fundername: NIBIB NIH HHS
  grantid: R56 EB033332
– fundername: NCI NIH HHS
  grantid: R01 CA272991
– fundername: NCI NIH HHS
  grantid: R01 CA215718
GroupedDBID 53G
AAGCD
AAJIO
AATNI
ABHWH
ABVAM
ACGFS
ACHIP
AFYNE
AKPSB
ALMA_UNASSIGNED_HOLDINGS
AOAED
ASPBG
ATQHT
CJUJL
CRLBU
EBS
IJHAN
IOP
IZVLO
KOT
N5L
O3W
PJBAE
RIN
ROL
RPA
TSCCA
ABJNI
CGR
CUY
CVF
ECM
EIF
NPM
7X8
ADEQX
AEINN
ID FETCH-LOGICAL-c466t-6c9c4c1a5cf987dc72fe97e36d26b97f9832d5c39e0248fe0df18e3089ea19493
IEDL.DBID O3W
ISICitedReferencesCount 6
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000985662000001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 2057-1976
IngestDate Wed Oct 01 13:38:41 EDT 2025
Thu Jan 30 12:29:35 EST 2025
Wed Aug 21 03:33:43 EDT 2024
Wed May 17 01:26:53 EDT 2023
IsDoiOpenAccess true
IsOpenAccess true
IsPeerReviewed true
IsScholarly true
Issue 4
Keywords signal denoising
Bragg peak
deep learning
stack auto-encoder
protoacoustic
Language English
License Original content from this work may be used under the terms of the Creative Commons Attribution 4.0 licence. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Creative Commons Attribution license.
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c466t-6c9c4c1a5cf987dc72fe97e36d26b97f9832d5c39e0248fe0df18e3089ea19493
Notes BPEX-103281.R1
ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ORCID 0000-0003-1618-2016
0000-0001-9023-5855
OpenAccessLink https://iopscience.iop.org/article/10.1088/2057-1976/acd257
PMID 37141867
PQID 2810923426
PQPubID 23479
PageCount 13
ParticipantIDs proquest_miscellaneous_2810923426
pubmed_primary_37141867
iop_journals_10_1088_2057_1976_acd257
PublicationCentury 2000
PublicationDate 20230512
PublicationDateYYYYMMDD 2023-05-12
PublicationDate_xml – month: 5
  year: 2023
  text: 20230512
  day: 12
PublicationDecade 2020
PublicationPlace England
PublicationPlace_xml – name: England
PublicationTitle Biomedical physics & engineering express
PublicationTitleAbbrev BPEX
PublicationTitleAlternate Biomed. Phys. Eng. Express
PublicationYear 2023
Publisher IOP Publishing
Publisher_xml – name: IOP Publishing
SSID ssj0001626695
Score 2.3072329
Snippet Proton therapy is a type of radiation therapy that can provide better dose distribution compared to photon therapy by delivering most of the energy at the end...
SourceID proquest
pubmed
iop
SourceType Aggregation Database
Index Database
Enrichment Source
Publisher
SubjectTerms Bragg peak
Deep Learning
protoacoustic
Proton Therapy - methods
Protons
Radiotherapy Dosage
Radiotherapy Planning, Computer-Assisted - methods
signal denoising
stack auto-encoder
Title Deep learning-based protoacoustic signal denoising for proton range verification
URI https://iopscience.iop.org/article/10.1088/2057-1976/acd257
https://www.ncbi.nlm.nih.gov/pubmed/37141867
https://www.proquest.com/docview/2810923426
Volume 9
WOSCitedRecordID wos000985662000001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV1LT8MwDLa2wYELD_EajylIcCysTdok4oSAiQOMHUDsFjUvtEtXbQPx83HaAkICCYlLVVVOG9mu_SV2bIBj9FCS9rWNcIHsI8aNjXTudOS1YSznLLO-6lpyy4dDMR7LUQvOP8_CTMvG9J_ibV0ouGZhkxAncLmecnw1z85yY1Hj2rBEBbpxVOZ7-vS1wYJQPZNpE5r8aSC6E_zG79CycjGDtX9Nbh1WG2RJLmrSDWi5YhNGV86VpGkN8RwFn2VJqM0wRUtYNfIiIYMDx6H9mU7CxgFBGFuTFGQWjh4Q1PaQUFTJcAseB9cPlzdR00QhMizLFlFmpGEmzlPjpeDW8MQ7yR3NbJJpyfEhTWxqqHShupl3fetj4WhfSJej_CTdhk4xLdwuEG2s8dTZVGrEISwWNnEB8KSMJtrbpAsnyBbV_ARzVcW3hVCBJyrwRNU86QL5RqdL96akYgrhZUizK63vwtGHUBTqewhi5IVDvqhExH0EpQgsurBTS0uVdWEOFaoPhgJ9e3-cyD6shO7xIRkgTg6gs5i9uENYNq-LyXzWgzYfC7wOR3e9SrXeAa9lzyE
linkProvider IOP Publishing
linkToPdf http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV1Na9wwEB2atIRc2oa0ifuRKpAe3V1bsiUdS9MlpWGzh4TuTVhfYS-22d2E_PzO2Aql0EKhN2Mk27wZa56k0RuAM4xQmk-tz3GCHHMhnc9tE2werROikaL2cahacinnc7Vc6kWqczqchen6NPR_wstRKHiEMCXEKZyuVxIfLetJ4zx63KT3cQeekk4JufUV__FrkQXpeq2rtD35p84YUvA9f6eXQ5iZvfjvD3wJzxPDZJ_H5gfwJLSHsDgPoWepRMRtTrHLM9Jo6HBEHAp6McrkwH44DnUrWkBgSGfHJi1b0xEEhl5PiUWDLV_Bzezr9ZeLPBVTyJ2o621eO-2EK5rKRa2kd7KMQcvAa1_WVku8yUtfOa4DqZzFMPWxUIFPlQ4N2lHz17Dbdm04Bmadd5EHX2mLfEQUypeBiE8leGmjLzP4iNCY9DNszLDPrZQhXAzhYkZcMmC_tbN9eDDaCIM0k9LtELYMTh8NY9DvaTOjaQPiYkpVTJGcIsHI4Gi0mOlHgQ5DKoQk1PfmHz_kA-wtzmfm8tv8-1vYp4LylB9QlO9gd7u-C-_hmbvfrjbrk8G7fgKKg9KV
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Deep+learning-based+protoacoustic+signal+denoising+for+proton+range+verification&rft.jtitle=Biomedical+physics+%26+engineering+express&rft.au=Wang%2C+Jing&rft.au=Sohn%2C+James+J&rft.au=Lei%2C+Yang&rft.au=Nie%2C+Wei&rft.date=2023-05-12&rft.pub=IOP+Publishing&rft.eissn=2057-1976&rft.volume=9&rft.issue=4&rft_id=info:doi/10.1088%2F2057-1976%2Facd257&rft.externalDocID=bpexacd257
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2057-1976&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2057-1976&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2057-1976&client=summon