Migration of Deep Learning Models Across Ultrasound Scanners

A transfer function approach has recently proven effective for calibrating deep learning (DL) algorithms in quantitative ultrasound (QUS), addressing data shifts at both the acquisition and machine levels. Expanding on this approach, we develop a strategy to acquire the functionality of a DL model f...

Celý popis

Uložené v:

Podrobná bibliografia
Vydané v:	IEEE transactions on biomedical engineering Ročník 72; číslo 11; s. 3210 - 3220
Hlavní autori:	Soylu, Ufuk, Chandrasekeran, Varun, Czarnota, Gregory J., Oelze, Michael L.
Médium:	Journal Article
Jazyk:	English
Vydavateľské údaje:	United States IEEE 01.11.2025 The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Predmet:	Adaptation models Algorithms Availability Biological system modeling Black boxes Calibration Closed box data mismatch Data models Decision making Deep Learning deep model security Humans Image Processing, Computer-Assisted - methods Information processing Machine learning Mental task performance Noise measurement Scanners Security Tissue characterization transfer function Transfer functions Ultrasonic imaging Ultrasonography - instrumentation Ultrasonography - methods Ultrasound ultrasound imaging
ISSN:	0018-9294, 1558-2531, 1558-2531
On-line prístup:	Získať plný text
Tagy:	Pridať tag Žiadne tagy, Buďte prvý, kto otaguje tento záznam!

Abstract	A transfer function approach has recently proven effective for calibrating deep learning (DL) algorithms in quantitative ultrasound (QUS), addressing data shifts at both the acquisition and machine levels. Expanding on this approach, we develop a strategy to acquire the functionality of a DL model from one ultrasound machine and implement it on another in a black-box setting, in the context of QUS. This demonstrates the ease with which the functionality of a DL model can be transferred between machines. While the proposed approach can also assist regulatory bodies in comparing and approving DL models, it also highlights the security risks associated with deploying such models in a commercial scanner for clinical use. The method is a black-box unsupervised domain adaptation technique that integrates the transfer function approach with an iterative schema. It does not utilize any information related to model internals but it solely relies on the availability of an input-output interface. Additionally, we assume the availability of unlabeled data from a testing machine. This scenario could become relevant as companies begin deploying their DL functionalities for clinical use. In the experiments, we used a SonixOne and a Verasonics machine. The model was trained on SonixOne data, and its functionality was then transferred to the Verasonics machine. The proposed method successfully transferred the functionality to the Verasonics machine, achieving a remarkable 98 percent classification accuracy in a binary decision task. This study underscores the need to establish security measures prior to deploying DL models in clinical settings.
AbstractList	A transfer function approach has recently proven effective for calibrating deep learning (DL) algorithms in quantitative ultrasound (QUS), addressing data shifts at both the acquisition and machine levels. Expanding on this approach, we develop a strategy to acquire the functionality of a DL model from one ultrasound machine and implement it on another in a black-box setting, in the context of QUS. This demonstrates the ease with which the functionality of a DL model can be transferred between machines. While the proposed approach can also assist regulatory bodies in comparing and approving DL models, it also highlights the security risks associated with deploying such models in a commercial scanner for clinical use. The method is a black-box unsupervised domain adaptation technique that integrates the transfer function approach with an iterative schema. It does not utilize any information related to model internals but it solely relies on the availability of an input-output interface. Additionally, we assume the availability of unlabeled data from a testing machine. This scenario could become relevant as companies begin deploying their DL functionalities for clinical use. In the experiments, we used a SonixOne and a Verasonics machine. The model was trained on SonixOne data, and its functionality was then transferred to the Verasonics machine. The proposed method successfully transferred the functionality to the Verasonics machine, achieving a remarkable 98 percent classification accuracy in a binary decision task. This study underscores the need to establish security measures prior to deploying DL models in clinical settings.A transfer function approach has recently proven effective for calibrating deep learning (DL) algorithms in quantitative ultrasound (QUS), addressing data shifts at both the acquisition and machine levels. Expanding on this approach, we develop a strategy to acquire the functionality of a DL model from one ultrasound machine and implement it on another in a black-box setting, in the context of QUS. This demonstrates the ease with which the functionality of a DL model can be transferred between machines. While the proposed approach can also assist regulatory bodies in comparing and approving DL models, it also highlights the security risks associated with deploying such models in a commercial scanner for clinical use. The method is a black-box unsupervised domain adaptation technique that integrates the transfer function approach with an iterative schema. It does not utilize any information related to model internals but it solely relies on the availability of an input-output interface. Additionally, we assume the availability of unlabeled data from a testing machine. This scenario could become relevant as companies begin deploying their DL functionalities for clinical use. In the experiments, we used a SonixOne and a Verasonics machine. The model was trained on SonixOne data, and its functionality was then transferred to the Verasonics machine. The proposed method successfully transferred the functionality to the Verasonics machine, achieving a remarkable 98 percent classification accuracy in a binary decision task. This study underscores the need to establish security measures prior to deploying DL models in clinical settings. A transfer function approach has recently proven effective for calibrating deep learning (DL) algorithms in quantitative ultrasound (QUS), addressing data shifts at both the acquisition and machine levels. Expanding on this approach, we develop a strategy to acquire the functionality of a DL model from one ultrasound machine and implement it on another in a black-box setting, in the context of QUS. This demonstrates the ease with which the functionality of a DL model can be transferred between machines. While the proposed approach can also assist regulatory bodies in comparing and approving DL models, it also highlights the security risks associated with deploying such models in a commercial scanner for clinical use. The method is a black-box unsupervised domain adaptation technique that integrates the transfer function approach with an iterative schema. It does not utilize any information related to model internals but it solely relies on the availability of an input-output interface. Additionally, we assume the availability of unlabeled data from a testing machine. This scenario could become relevant as companies begin deploying their DL functionalities for clinical use. In the experiments, we used a SonixOne and a Verasonics machine. The model was trained on SonixOne data, and its functionality was then transferred to the Verasonics machine. The proposed method successfully transferred the functionality to the Verasonics machine, achieving a remarkable 98 percent classification accuracy in a binary decision task. This study underscores the need to establish security measures prior to deploying DL models in clinical settings.
Author	Czarnota, Gregory J. Chandrasekeran, Varun Oelze, Michael L. Soylu, Ufuk
Author_xml	– sequence: 1 givenname: Ufuk orcidid: 0000-0002-0513-4572 surname: Soylu fullname: Soylu, Ufuk email: usoylu2@illinois.edu organization: Department of Electrical and Computer Engineering, University of Illinois at Urbana-Champaign, USA – sequence: 2 givenname: Varun orcidid: 0000-0003-0661-6378 surname: Chandrasekeran fullname: Chandrasekeran, Varun organization: Department of Electrical and Computer Engineering, University of Illinois at Urbana-Champaign, USA – sequence: 3 givenname: Gregory J. surname: Czarnota fullname: Czarnota, Gregory J. organization: Sunnybrook Health Sciences Centre, Canada – sequence: 4 givenname: Michael L. orcidid: 0000-0002-3099-7425 surname: Oelze fullname: Oelze, Michael L. email: oelze@illinois.edu organization: Department of Electrical and Computer Engineering, University of Illinois at Urbana-Champaign,, Urbana, IL, USA
BackLink	https://www.ncbi.nlm.nih.gov/pubmed/40279236$$D View this record in MEDLINE/PubMed
BookMark	eNpdkMtKA0EQRRuJmBj9AEFkwI2bif1-gJv4FhJcmKybzkyNTJh0x-7Mwr93YqKIq6Lg3OLWOUY9HzwgdEbwiBBsrme304cRxVSMmJBcSHWABkQInVPBSA8NMCY6N9TwPjpOadmtXHN5hPocU2UokwN0M63fo9vUwWehyu4B1tkEXPS1f8-moYQmZeMihpSyebOJLoXWl9lb4byHmE7QYeWaBKf7OUTzx4fZ3XM-eX16uRtP8oJxtck5XrCi0loUzilBSuGMZAtOBRGgCwlCywJzAQpL7MTClY5pTSsuGNWmkpgN0dXu7jqGjxbSxq7qVEDTOA-hTZYRw41SUuoOvfyHLkMbfdfOMiqZYZ0m0VEXe6pdrKC061ivXPy0P146gOyA798jVL8IwXbr3m7d2617u3ffZc53mRoA_vBdM6U0-wKIuHyL
CODEN	IEBEAX
Cites_doi	10.1109/TUFFC.2021.3075912 10.7863/jum.2005.24.9.1235 10.1038/s41598-022-06100-2 10.1109/CVPR.2016.90 10.1109/TUFFC.2023.3263119 10.1109/ICASSP.2018.8461575 10.1109/TUFFC.2024.3384815 10.1007/978-3-030-87234-2_2 10.1109/IUS52206.2021.9593665 10.1109/CVPR.2017.243 10.5244/C.35.114 10.1109/IUS52206.2021.9593856 10.1007/978-3-031-16446-0_74 10.1016/j.ultras.2021.106682 10.1109/TUFFC.2024.3357438 10.1109/TMI.2020.2970867 10.1109/TUFFC.2020.3026598 10.1109/TUFFC.2021.3094849 10.1609/aaai.v38i21.30343 10.1007/s11263-024-02181-w 10.1109/ULTSYM.2019.8925666 10.1016/j.ultrasmedbio.2020.10.025 10.1038/s41467-020-17478-w 10.1109/JPROC.2019.2932116 10.1109/TUFFC.2023.3245988 10.1148/radiol.2020191160 10.1177/016173461003200104 10.1118/1.594483 10.1016/S0301-5629(98)00013-1 10.1002/jum.15693 10.1148/radiol.221510 10.1007/978-3-030-59713-9_40 10.1109/TUFFC.2022.3144685
ContentType	Journal Article
Copyright	Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2025
Copyright_xml	– notice: Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2025
DBID	97E ESBDL RIA RIE AAYXX CITATION CGR CUY CVF ECM EIF NPM 7QF 7QO 7QQ 7SC 7SE 7SP 7SR 7TA 7TB 7U5 8BQ 8FD F28 FR3 H8D JG9 JQ2 KR7 L7M L~C L~D P64 7X8
DOI	10.1109/TBME.2025.3564567
DatabaseName	IEEE All-Society Periodicals Package (ASPP) 2005–Present IEEE Xplore Open Access Journals IEEE All-Society Periodicals Package (ASPP) 1998–Present IEEE Electronic Library (IEL) CrossRef Medline MEDLINE MEDLINE (Ovid) MEDLINE MEDLINE PubMed Aluminium Industry Abstracts Biotechnology Research Abstracts Ceramic Abstracts Computer and Information Systems Abstracts Corrosion Abstracts Electronics & Communications Abstracts Engineered Materials Abstracts Materials Business File Mechanical & Transportation Engineering Abstracts Solid State and Superconductivity Abstracts METADEX Technology Research Database ANTE: Abstracts in New Technology & Engineering Engineering Research Database Aerospace Database Materials Research Database ProQuest Computer Science Collection Civil Engineering Abstracts Advanced Technologies Database with Aerospace Computer and Information Systems Abstracts Academic Computer and Information Systems Abstracts Professional Biotechnology and BioEngineering Abstracts MEDLINE - Academic
DatabaseTitle	CrossRef MEDLINE Medline Complete MEDLINE with Full Text PubMed MEDLINE (Ovid) Materials Research Database Civil Engineering Abstracts Aluminium Industry Abstracts Technology Research Database Computer and Information Systems Abstracts – Academic Mechanical & Transportation Engineering Abstracts Electronics & Communications Abstracts ProQuest Computer Science Collection Computer and Information Systems Abstracts Ceramic Abstracts Materials Business File METADEX Biotechnology and BioEngineering Abstracts Computer and Information Systems Abstracts Professional Aerospace Database Engineered Materials Abstracts Biotechnology Research Abstracts Solid State and Superconductivity Abstracts Engineering Research Database Corrosion Abstracts Advanced Technologies Database with Aerospace ANTE: Abstracts in New Technology & Engineering MEDLINE - Academic
DatabaseTitleList	MEDLINE - Academic Materials Research Database 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: RIE name: IEEE Electronic Library (IEL) url: https://ieeexplore.ieee.org/ sourceTypes: Publisher – sequence: 3 dbid: 7X8 name: MEDLINE - Academic url: https://search.proquest.com/medline sourceTypes: Aggregation Database
DeliveryMethod	fulltext_linktorsrc
Discipline	Medicine Engineering
EISSN	1558-2531
EndPage	3220
ExternalDocumentID	40279236 10_1109_TBME_2025_3564567 10977778
Genre	orig-research Journal Article
GrantInformation_xml	– fundername: National Institutes of Health grantid: R01CA251939; R01CA273700 funderid: 10.13039/100000002 – fundername: NCI NIH HHS grantid: R01 CA251939 – fundername: NCI NIH HHS grantid: R01 CA273700
GroupedDBID	--- -~X .55 .DC .GJ 0R~ 29I 4.4 53G 5GY 5RE 5VS 6IF 6IK 85S 97E AAJGR AASAJ AAWTH AAYJJ ABAZT ABJNI ABQJQ ABVLG ACGFO ACGFS ACIWK ACKIV ACNCT ACPRK AENEX AETIX AFFNX AFRAH AGQYO AGSQL AHBIQ AI. AIBXA AKJIK AKQYR ALLEH ALMA_UNASSIGNED_HOLDINGS ASUFR ATWAV BEFXN BFFAM BGNUA BKEBE BPEOZ CS3 DU5 EBS EJD ESBDL F5P HZ~ H~9 IAAWW IBMZZ ICLAB IDIHD IFIPE IFJZH IPLJI JAVBF LAI MS~ O9- OCL P2P RIA RIE RNS TAE TN5 VH1 VJK X7M ZGI ZXP AAYXX CITATION CGR CUY CVF ECM EIF NPM 7QF 7QO 7QQ 7SC 7SE 7SP 7SR 7TA 7TB 7U5 8BQ 8FD F28 FR3 H8D JG9 JQ2 KR7 L7M L~C L~D P64 7X8
ID	FETCH-LOGICAL-c347t-40b3cf885caa751d5a963b42515e8c6e586c045e7060a5bada3882f453289f603
IEDL.DBID	RIE
ISICitedReferencesCount	0
ISICitedReferencesURI	http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001600066000001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN	0018-9294 1558-2531
IngestDate	Fri Oct 03 00:10:48 EDT 2025 Sat Nov 15 15:52:25 EST 2025 Sat Oct 25 06:00:16 EDT 2025 Sat Nov 29 06:50:28 EST 2025 Wed Nov 19 08:27:10 EST 2025
IsDoiOpenAccess	true
IsOpenAccess	true
IsPeerReviewed	true
IsScholarly	true
Issue	11
Language	English
License	https://creativecommons.org/licenses/by/4.0/legalcode
LinkModel	DirectLink
MergedId	FETCHMERGED-LOGICAL-c347t-40b3cf885caa751d5a963b42515e8c6e586c045e7060a5bada3882f453289f603
Notes	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14 content type line 23
ORCID	0000-0003-0661-6378 0000-0002-3099-7425 0000-0002-0513-4572
OpenAccessLink	https://ieeexplore.ieee.org/document/10977778
PMID	40279236
PQID	3263934565
PQPubID	85474
PageCount	11
ParticipantIDs	crossref_primary_10_1109_TBME_2025_3564567 ieee_primary_10977778 pubmed_primary_40279236 proquest_journals_3263934565 proquest_miscellaneous_3194977668
PublicationCentury	2000
PublicationDate	2025-11-01
PublicationDateYYYYMMDD	2025-11-01
PublicationDate_xml	– month: 11 year: 2025 text: 2025-11-01 day: 01
PublicationDecade	2020
PublicationPlace	United States
PublicationPlace_xml	– name: United States – name: New York
PublicationTitle	IEEE transactions on biomedical engineering
PublicationTitleAbbrev	TBME
PublicationTitleAlternate	IEEE Trans Biomed Eng
PublicationYear	2025
Publisher	IEEE The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Publisher_xml	– name: IEEE – name: The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
References	ref13 ref12 ref15 ref14 ref31 ref30 Paszke (ref35) 2019 ref11 ref33 ref10 ref32 ref2 ref1 ref17 ref16 ref19 ref18 ref24 ref23 ref26 ref25 ref20 ref22 ref21 ref28 ref27 ref29 ref8 ref7 ref9 Chandrasekaran (ref36) 2020 ref4 ref3 ref6 Kingma (ref34) 2015 ref5
References_xml	– ident: ref12 doi: 10.1109/TUFFC.2021.3075912 – ident: ref28 doi: 10.7863/jum.2005.24.9.1235 – ident: ref21 doi: 10.1038/s41598-022-06100-2 – ident: ref33 doi: 10.1109/CVPR.2016.90 – ident: ref7 doi: 10.1109/TUFFC.2023.3263119 – ident: ref4 doi: 10.1109/ICASSP.2018.8461575 – ident: ref23 doi: 10.1109/TUFFC.2024.3384815 – ident: ref11 doi: 10.1007/978-3-030-87234-2_2 – ident: ref5 doi: 10.1109/IUS52206.2021.9593665 – ident: ref32 doi: 10.1109/CVPR.2017.243 – ident: ref27 doi: 10.5244/C.35.114 – ident: ref13 doi: 10.1109/IUS52206.2021.9593856 – ident: ref14 doi: 10.1007/978-3-031-16446-0_74 – ident: ref16 doi: 10.1016/j.ultras.2021.106682 – ident: ref15 doi: 10.1109/TUFFC.2024.3357438 – ident: ref24 doi: 10.1109/TMI.2020.2970867 – ident: ref2 doi: 10.1109/TUFFC.2020.3026598 – ident: ref3 doi: 10.1109/TUFFC.2021.3094849 – ident: ref25 doi: 10.1609/aaai.v38i21.30343 – ident: ref26 doi: 10.1007/s11263-024-02181-w – ident: ref19 doi: 10.1109/ULTSYM.2019.8925666 – ident: ref20 doi: 10.1016/j.ultrasmedbio.2020.10.025 – ident: ref6 doi: 10.1038/s41467-020-17478-w – start-page: 1309 volume-title: Proc. 29th USENIX Secur. Symp. year: 2020 ident: ref36 article-title: Exploring connections between active learning and model extraction – ident: ref1 doi: 10.1109/JPROC.2019.2932116 – volume-title: Proc. Int. Conf. Learn. Representations year: 2015 ident: ref34 article-title: Adam: A method for stochastic optimization – start-page: 8026 volume-title: Proc. Adv. Neural Inf. Process. Syst. year: 2019 ident: ref35 article-title: PyTorch: An imperative style, high-performance deep learning library – ident: ref8 doi: 10.1109/TUFFC.2023.3245988 – ident: ref18 doi: 10.1148/radiol.2020191160 – ident: ref30 doi: 10.1177/016173461003200104 – ident: ref31 doi: 10.1118/1.594483 – ident: ref29 doi: 10.1016/S0301-5629(98)00013-1 – ident: ref17 doi: 10.1002/jum.15693 – ident: ref22 doi: 10.1148/radiol.221510 – ident: ref9 doi: 10.1007/978-3-030-59713-9_40 – ident: ref10 doi: 10.1109/TUFFC.2022.3144685
SSID	ssj0014846
Score	2.481593
Snippet	A transfer function approach has recently proven effective for calibrating deep learning (DL) algorithms in quantitative ultrasound (QUS), addressing data...
SourceID	proquest pubmed crossref ieee
SourceType	Aggregation Database Index Database Publisher
StartPage	3210
SubjectTerms	Adaptation models Algorithms Availability Biological system modeling Black boxes Calibration Closed box data mismatch Data models Decision making Deep Learning deep model security Humans Image Processing, Computer-Assisted - methods Information processing Machine learning Mental task performance Noise measurement Scanners Security Tissue characterization transfer function Transfer functions Ultrasonic imaging Ultrasonography - instrumentation Ultrasonography - methods Ultrasound ultrasound imaging
Title	Migration of Deep Learning Models Across Ultrasound Scanners
URI	https://ieeexplore.ieee.org/document/10977778 https://www.ncbi.nlm.nih.gov/pubmed/40279236 https://www.proquest.com/docview/3263934565 https://www.proquest.com/docview/3194977668
Volume	72
WOSCitedRecordID	wos001600066000001&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
journalDatabaseRights	– providerCode: PRVIEE databaseName: IEEE Electronic Library (IEL) customDbUrl: eissn: 1558-2531 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0014846 issn: 0018-9294 databaseCode: RIE dateStart: 19640101 isFulltext: true titleUrlDefault: https://ieeexplore.ieee.org/ providerName: IEEE
link	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1NS8NAEB2siOjBj1q1WmUFT0LbpNndbMBL1YqXFkGF3sImmYggqfTD3-_MJi166MGcAtlsln07m_cymRmAa2W1Z_2Qdj-ZyLZM0W-bxEeyqyzUyBQAPVdsIhyNzHgcPVfB6i4WBhHdz2fY4VPny88m6YI_lXXZW0qHqUEtDHUZrLVyGUhTRuV4PllwL5KVC5Pu6b7eDQckBXuqE3DyFM2V90g3ceo8_ed95AqsrOea7p3zuP_P0R7AXkUuRb9cDYewgUUddn-lHKzD9rByph_B7fDjvcRfTHLxgPglqmyr74JLpH3ORN8NWrx9zqd2xgWYxAtBURBlbMDb4-D1_qldFVNop4EM56QTkyDNjVGptaHyM2XJ9BKyWF-hSTUqo1Oid8jZdKxKbGYDIt-5VAFJslx7wTFsFpMCT0H4IYsWYzOSI7JniWPShp9HeU5sR2WomnCznNL4q8yZETut4UUxQxEzFHEFRRMaPHW_Gpaz1oTWEoW4MqtZTFwziAImoU24Wl0mg2Avhy1wsqA2fiSpC62pi5MSvVXnS9DP1jz0HHZ4bGWsYQs259MFXsBW-j3_mE0vadWNzaVbdT-sss5f
linkProvider	IEEE
linkToHtml	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3fT9swED4BQ8AexoBuK-uYJ_GEVJo0tuNIvDBWBIJWSCsSb5GTXKpKKEX9wd_PneNW8MADeYoUx7H8-Zzvcrn7AI6V1YENY9r9ZCbbMsewbbIQya6KWCNTAAyc2EQ8GJiHh-TOJ6u7XBhEdD-f4Smfulh-MckX_Kmsw9FSOsw6fGLpLJ-utQoaSFPn5QQh2XA3kT6ISXd1hn_7PXIGu-o04vIpmrX3yHPi4nn6zRvJSay8zzbdW-dy94Pj_QpfPL0U5_V62IM1rPbh86uig_uw1ffh9AM4649H9QoQk1L8Q3wSvt7qSLBI2uNMnLtBi_vH-dTOWIJJ_CcwKiKNDbi_7A0vrtpeTqGdRzKek6eYRXlpjMqtjVVYKEvGl5HNhgpNrlEZnRPBQ66nY1VmCxsR_S6lisgpK3UQfYONalLhDxBhzG6LsQU5JLJriWXSll8mZUl8RxWomnCynNL0qa6akTpvI0hShiJlKFIPRRMaPHWvGtaz1oTWEoXUG9YsJbYZJRHT0Cb8WV0mk-A4h61wsqA2YSKpC62pi-81eqvOl6AfvvPQ37B9NezfprfXg5ufsMPjrDMPW7Axny7wF2zmz_PxbHrk1t4LveDQwA
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=Migration+of+Deep+Learning+Models+Across+Ultrasound+Scanners&rft.jtitle=IEEE+transactions+on+biomedical+engineering&rft.au=Soylu%2C+Ufuk&rft.au=Chandrasekeran%2C+Varun&rft.au=Czarnota%2C+Gregory+J&rft.au=Oelze%2C+Michael+L&rft.date=2025-11-01&rft.eissn=1558-2531&rft.volume=72&rft.issue=11&rft.spage=3210&rft_id=info:doi/10.1109%2FTBME.2025.3564567&rft_id=info%3Apmid%2F40279236&rft.externalDocID=40279236
thumbnail_l	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0018-9294&client=summon
thumbnail_m	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0018-9294&client=summon
thumbnail_s	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0018-9294&client=summon