An Energy-Efficient Accelerator for Medical Image Reconstruction From Implicit Neural Representation

This work presents an energy-efficient accelerator for medical image reconstruction from implicit neural representation (INR). The accelerator implements an INR-based algorithm to deliver high-quality medical image reconstruction with arbitrary resolution from a compact implicit format. In particula...

Celý popis

Uložené v:

Podrobná bibliografia
Vydané v:	IEEE transactions on circuits and systems. I, Regular papers Ročník 70; číslo 4; s. 1625 - 1638
Hlavní autori:	Rao, Chaolin, Wu, Qing, Zhou, Pingqiang, Yu, Jingyi, Zhang, Yuyao, Lou, Xin
Médium:	Journal Article
Jazyk:	English
Vydavateľské údaje:	New York IEEE 01.04.2023 The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Predmet:	accelerator Algorithms Biological neural networks Central processing units Computation Computational modeling Computer architecture CPUs Energy efficiency energy-efficient Field programmable gate arrays Frames per second Graphics processing units Image coding Image quality Image reconstruction Image resolution implicit neural representation (INR) Medical diagnostic imaging Medical image reconstruction Medical imaging Performance evaluation Power consumption Representations Three-dimensional displays
ISSN:	1549-8328, 1558-0806
On-line prístup:	Získať plný text
Tagy:	Pridať tag Žiadne tagy, Buďte prvý, kto otaguje tento záznam!

Abstract	This work presents an energy-efficient accelerator for medical image reconstruction from implicit neural representation (INR). The accelerator implements an INR-based algorithm to deliver high-quality medical image reconstruction with arbitrary resolution from a compact implicit format. In particular, we propose a dedicated hardware architecture based on an optimized computation flow for the INR-based reconstruction algorithm, which co-designs data reuse and computation load. The proposed architecture takes in the coordinate of the intersection of three scans and outputs all the voxel intensities, minimizing the data movement between on-chip and off-chip. To validate the proposed accelerator, we build a proof-of-concept prototype demonstration system using field programmable gate array (FPGA). We also map our design to 40nm CMOS technology to measure the performance of the proposed accelerator. The implementation results show that, running at 400MHz, the proposed accelerator is capable of processing medical images with <inline-formula> <tex-math notation="LaTeX">256\times 256 </tex-math></inline-formula> resolution in real-time at 26.3 frames per second (FPS), with a power consumption of only 795 mW. Comparison results show that the performance, as well as the energy efficiency of the proposed accelerator, outperforms the central processing unit (CPU)-based and graphic processing unit (GPU)-based implementations.
AbstractList	This work presents an energy-efficient accelerator for medical image reconstruction from implicit neural representation (INR). The accelerator implements an INR-based algorithm to deliver high-quality medical image reconstruction with arbitrary resolution from a compact implicit format. In particular, we propose a dedicated hardware architecture based on an optimized computation flow for the INR-based reconstruction algorithm, which co-designs data reuse and computation load. The proposed architecture takes in the coordinate of the intersection of three scans and outputs all the voxel intensities, minimizing the data movement between on-chip and off-chip. To validate the proposed accelerator, we build a proof-of-concept prototype demonstration system using field programmable gate array (FPGA). We also map our design to 40nm CMOS technology to measure the performance of the proposed accelerator. The implementation results show that, running at 400MHz, the proposed accelerator is capable of processing medical images with [Formula Omitted] resolution in real-time at 26.3 frames per second (FPS), with a power consumption of only 795 mW. Comparison results show that the performance, as well as the energy efficiency of the proposed accelerator, outperforms the central processing unit (CPU)-based and graphic processing unit (GPU)-based implementations. This work presents an energy-efficient accelerator for medical image reconstruction from implicit neural representation (INR). The accelerator implements an INR-based algorithm to deliver high-quality medical image reconstruction with arbitrary resolution from a compact implicit format. In particular, we propose a dedicated hardware architecture based on an optimized computation flow for the INR-based reconstruction algorithm, which co-designs data reuse and computation load. The proposed architecture takes in the coordinate of the intersection of three scans and outputs all the voxel intensities, minimizing the data movement between on-chip and off-chip. To validate the proposed accelerator, we build a proof-of-concept prototype demonstration system using field programmable gate array (FPGA). We also map our design to 40nm CMOS technology to measure the performance of the proposed accelerator. The implementation results show that, running at 400MHz, the proposed accelerator is capable of processing medical images with <inline-formula> <tex-math notation="LaTeX">256\times 256 </tex-math></inline-formula> resolution in real-time at 26.3 frames per second (FPS), with a power consumption of only 795 mW. Comparison results show that the performance, as well as the energy efficiency of the proposed accelerator, outperforms the central processing unit (CPU)-based and graphic processing unit (GPU)-based implementations.
Author	Zhang, Yuyao Lou, Xin Rao, Chaolin Zhou, Pingqiang Yu, Jingyi Wu, Qing
Author_xml	– sequence: 1 givenname: Chaolin orcidid: 0000-0003-2172-5361 surname: Rao fullname: Rao, Chaolin email: raochl@shanghaitech.edu.cn organization: School of Information Science and Technology, ShanghaiTech University, Shanghai, China – sequence: 2 givenname: Qing orcidid: 0000-0001-9320-0332 surname: Wu fullname: Wu, Qing email: wuqing@shanghaitech.edu.cn organization: School of Information Science and Technology, ShanghaiTech University, Shanghai, China – sequence: 3 givenname: Pingqiang orcidid: 0000-0001-9515-9302 surname: Zhou fullname: Zhou, Pingqiang email: zhoupq@shanghaitech.edu.cn organization: School of Information Science and Technology, ShanghaiTech University, Shanghai, China – sequence: 4 givenname: Jingyi orcidid: 0000-0001-9198-6853 surname: Yu fullname: Yu, Jingyi email: yujingyi@shanghaitech.edu.cn organization: School of Information Science and Technology, ShanghaiTech University, Shanghai, China – sequence: 5 givenname: Yuyao orcidid: 0000-0001-6706-4867 surname: Zhang fullname: Zhang, Yuyao email: zhangyy8@shanghaitech.edu.cn organization: School of Information Science and Technology, ShanghaiTech University, Shanghai, China – sequence: 6 givenname: Xin orcidid: 0000-0002-8499-5038 surname: Lou fullname: Lou, Xin email: louxin@shanghaitech.edu.cn organization: School of Information Science and Technology, ShanghaiTech University, Shanghai, China
BookMark	eNp9kE1LAzEQhoNUsK3-AMHDguetk4_9OpbSaqEq1Hpe0mRStmyTNbs99N-btT2IBw9DBuZ5JsM7IgPrLBJyT2FCKRRPm9nHcsKAsQlnnOYpvyJDmiR5DDmkg74XRZxzlt-QUdvuAVgBnA6JntpobtHvTvHcmEpVaLtoqhTW6GXnfGRCvaKulKyj5UHuMFqjcrbt_FF1lbPRwrtDmDR1kLvoDY8-kGtsPLZhl-yZW3JtZN3i3eUdk8_FfDN7iVfvz8vZdBUrVoguTsCAoQU1JktSuRWQCS04Mppp0KiMVlBQqXOjKFKdQJanORWoJM-k3hrBx-TxvLfx7uuIbVfu3dHb8GXJsoIzCLmkgcrOlPKubT2aMhz-c2fnZVWXFMo-0rKPtOwjLS-RBpP-MRtfHaQ__es8nJ0KEX_xAEKE-Tc9B4Uw
CODEN	ITCSCH
CitedBy_id	crossref_primary_10_1088_1361_6560_ad33b7 crossref_primary_10_1109_TMI_2025_3526452 crossref_primary_10_1109_TMI_2023_3342156 crossref_primary_10_1109_TCSI_2024_3488858 crossref_primary_10_1109_TCSI_2024_3413570 crossref_primary_10_1088_1361_6560_addfa5 crossref_primary_10_1109_TSUSC_2023_3325268 crossref_primary_10_1109_TVLSI_2024_3515113 crossref_primary_10_1145_3727257_3727259
Cites_doi	10.1109/MICRO.2016.7783723 10.1109/TMI.2019.2922960 10.1007/978-3-030-58452-8_24 10.1109/CVPR.2019.00459 10.1109/TIP.2019.2942510 10.1109/TCSVT.2006.876360 10.1109/TCSI.2017.2759803 10.1109/TCSVT.2018.2812781 10.1109/HPCA47549.2020.00030 10.1007/978-3-030-59713-9_14 10.1007/978-3-030-87231-1_7 10.1016/0893-6080(89)90020-8 10.1109/TCSI.2017.2757036 10.1111/cgf.14507 10.1109/TCSI.2022.3188899 10.1145/3474085.3475584 10.1109/TCSI.2020.3030663 10.1109/TVLSI.2017.2688340 10.1109/TMI.2020.3037187 10.1109/CVPR.2016.90 10.1007/978-3-031-19787-1_10 10.1109/AICAS51828.2021.9458511 10.1109/CVPR.2019.00025 10.1109/TCSVT.2011.2129350 10.1007/s003300101100 10.1109/TCSI.2018.2840092 10.1109/CVPR46437.2021.00852 10.1109/VLSIC.2018.8502333 10.1109/ISSCC.2017.7870353 10.1109/JSSC.2016.2616357 10.1109/TVLSI.2021.3135353 10.23919/VLSIT.2019.8776507
ContentType	Journal Article
Copyright	Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2023
Copyright_xml	– notice: Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2023
DBID	97E RIA RIE AAYXX CITATION 7SP 8FD L7M
DOI	10.1109/TCSI.2022.3231863
DatabaseName	IEEE All-Society Periodicals Package (ASPP) 2005–Present IEEE All-Society Periodicals Package (ASPP) 1998–Present IEEE Electronic Library (IEL) CrossRef Electronics & Communications Abstracts Technology Research Database Advanced Technologies Database with Aerospace
DatabaseTitle	CrossRef Technology Research Database Advanced Technologies Database with Aerospace Electronics & Communications Abstracts
DatabaseTitleList	Technology Research Database
Database_xml	– sequence: 1 dbid: RIE name: IEEE Explore url: https://ieeexplore.ieee.org/ sourceTypes: Publisher
DeliveryMethod	fulltext_linktorsrc
Discipline	Engineering
EISSN	1558-0806
EndPage	1638
ExternalDocumentID	10_1109_TCSI_2022_3231863 10004486
Genre	orig-research
GrantInformation_xml	– fundername: Shanghai Rising-Star Program grantid: 21QC1401400 – fundername: Central Guided Local Science and Technology Foundation of China grantid: YDZX20223100001001
GroupedDBID	0R~ 29I 4.4 5VS 6IK 97E AAJGR AARMG AASAJ AAWTH ABAZT ABQJQ ABVLG ACIWK AETIX AGQYO AGSQL AHBIQ AIBXA AKJIK AKQYR ALMA_UNASSIGNED_HOLDINGS ATWAV BEFXN BFFAM BGNUA BKEBE BPEOZ EBS EJD HZ~ H~9 IFIPE IPLJI JAVBF M43 O9- OCL PZZ RIA RIE RNS VJK AAYXX CITATION 7SP 8FD L7M
ID	FETCH-LOGICAL-c294t-50f0f191ff756ab4074d43e217d0decfdc091ad8fc1e1d50786814eca37adbf43
IEDL.DBID	RIE
ISICitedReferencesCount	11
ISICitedReferencesURI	http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000910584100001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN	1549-8328
IngestDate	Mon Jun 30 04:22:17 EDT 2025 Tue Nov 18 22:23:58 EST 2025 Sat Nov 29 06:23:57 EST 2025 Wed Aug 27 02:49:16 EDT 2025
IsPeerReviewed	true
IsScholarly	true
Issue	4
Language	English
License	https://ieeexplore.ieee.org/Xplorehelp/downloads/license-information/IEEE.html https://doi.org/10.15223/policy-029 https://doi.org/10.15223/policy-037
LinkModel	DirectLink
MergedId	FETCHMERGED-LOGICAL-c294t-50f0f191ff756ab4074d43e217d0decfdc091ad8fc1e1d50786814eca37adbf43
Notes	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14
ORCID	0000-0001-6706-4867 0000-0002-8499-5038 0000-0001-9320-0332 0000-0001-9515-9302 0000-0001-9198-6853 0000-0003-2172-5361
PQID	2793208636
PQPubID	85411
PageCount	14
ParticipantIDs	crossref_primary_10_1109_TCSI_2022_3231863 ieee_primary_10004486 crossref_citationtrail_10_1109_TCSI_2022_3231863 proquest_journals_2793208636
PublicationCentury	2000
PublicationDate	2023-04-01
PublicationDateYYYYMMDD	2023-04-01
PublicationDate_xml	– month: 04 year: 2023 text: 2023-04-01 day: 01
PublicationDecade	2020
PublicationPlace	New York
PublicationPlace_xml	– name: New York
PublicationTitle	IEEE transactions on circuits and systems. I, Regular papers
PublicationTitleAbbrev	TCSI
PublicationYear	2023
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 ref35 Tancik (ref7); 33 ref12 ref34 ref15 ref37 ref14 ref36 ref31 ref11 ref10 ref32 ref1 ref17 ref16 ref19 Rahaman (ref30) ref18 Sitzmann (ref6); 33 ref24 ref23 ref26 ref25 ref20 ref22 ref21 Mustra (ref2); 1 ref28 ref27 ref29 ref8 ref9 Wu (ref33) 2022 ref4 ref3 ref5
References_xml	– ident: ref28 doi: 10.1109/MICRO.2016.7783723 – ident: ref34 doi: 10.1109/TMI.2019.2922960 – ident: ref5 doi: 10.1007/978-3-030-58452-8_24 – ident: ref9 doi: 10.1109/CVPR.2019.00459 – volume: 1 start-page: 39 volume-title: Proc. 50th Int. Symp. ELMAR ident: ref2 article-title: Overview of the DICOM standard – ident: ref37 doi: 10.1109/TIP.2019.2942510 – start-page: 5301 volume-title: Proc. Int. Conf. Mach. Learn. ident: ref30 article-title: On the spectral bias of neural networks – ident: ref3 doi: 10.1109/TCSVT.2006.876360 – ident: ref22 doi: 10.1109/TCSI.2017.2759803 – ident: ref23 doi: 10.1109/TCSVT.2018.2812781 – ident: ref24 doi: 10.1109/HPCA47549.2020.00030 – ident: ref36 doi: 10.1007/978-3-030-59713-9_14 – ident: ref15 doi: 10.1007/978-3-030-87231-1_7 – ident: ref29 doi: 10.1016/0893-6080(89)90020-8 – ident: ref20 doi: 10.1109/TCSI.2017.2757036 – ident: ref10 doi: 10.1111/cgf.14507 – ident: ref13 doi: 10.1109/TCSI.2022.3188899 – volume: 33 start-page: 7537 volume-title: Proc. Adv. Neural Inf. Process. Syst. ident: ref7 article-title: Fourier features let networks learn high frequency functions in low dimensional domains – ident: ref16 doi: 10.1145/3474085.3475584 – ident: ref19 doi: 10.1109/TCSI.2020.3030663 – ident: ref18 doi: 10.1109/TVLSI.2017.2688340 – ident: ref35 doi: 10.1109/TMI.2020.3037187 – year: 2022 ident: ref33 article-title: Self-supervised coordinate projection network for sparse-view computed tomography publication-title: arXiv:2209.05483 – ident: ref32 doi: 10.1109/CVPR.2016.90 – ident: ref17 doi: 10.1007/978-3-031-19787-1_10 – ident: ref27 doi: 10.1109/AICAS51828.2021.9458511 – ident: ref8 doi: 10.1109/CVPR.2019.00025 – ident: ref4 doi: 10.1109/TCSVT.2011.2129350 – ident: ref1 doi: 10.1007/s003300101100 – ident: ref21 doi: 10.1109/TCSI.2018.2840092 – ident: ref14 doi: 10.1109/CVPR46437.2021.00852 – ident: ref26 doi: 10.1109/VLSIC.2018.8502333 – ident: ref11 doi: 10.1109/ISSCC.2017.7870353 – ident: ref12 doi: 10.1109/JSSC.2016.2616357 – ident: ref31 doi: 10.1109/TVLSI.2021.3135353 – volume: 33 start-page: 7462 volume-title: Proc. Adv. Neural Inf. Process. Syst. ident: ref6 article-title: Implicit neural representations with periodic activation functions – ident: ref25 doi: 10.23919/VLSIT.2019.8776507
SSID	ssj0029031
Score	2.4559798
Snippet	This work presents an energy-efficient accelerator for medical image reconstruction from implicit neural representation (INR). The accelerator implements an...
SourceID	proquest crossref ieee
SourceType	Aggregation Database Enrichment Source Index Database Publisher
StartPage	1625
SubjectTerms	accelerator Algorithms Biological neural networks Central processing units Computation Computational modeling Computer architecture CPUs Energy efficiency energy-efficient Field programmable gate arrays Frames per second Graphics processing units Image coding Image quality Image reconstruction Image resolution implicit neural representation (INR) Medical diagnostic imaging Medical image reconstruction Medical imaging Performance evaluation Power consumption Representations Three-dimensional displays
Title	An Energy-Efficient Accelerator for Medical Image Reconstruction From Implicit Neural Representation
URI	https://ieeexplore.ieee.org/document/10004486 https://www.proquest.com/docview/2793208636
Volume	70
WOSCitedRecordID	wos000910584100001&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 Explore customDbUrl: eissn: 1558-0806 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0029031 issn: 1549-8328 databaseCode: RIE dateStart: 20040101 isFulltext: true titleUrlDefault: https://ieeexplore.ieee.org/ providerName: IEEE
link	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LSwMxEA62eNCDz4rVKjl4EmJ3N-kmOZbSYkGK-MLbsk0yIOhWavX3O8mmPhAF97SHZFnyZSfz7cx8Q8hJj2sLUmcshYQz4coe0wqA5VBKLkuQELT07i7kZKLu7_VlLFYPtTDOuZB85s78bYjl25l59b_KummIP6q8QRpSyrpY64Nd6YTX4qhCM9ymKoYw00R3bwbXY6SCWXbG0Z1ROf92CIWuKj9McThfRpv_fLMtshEdSdqvkd8mK67aIetf5AV3ie1XdBhq-9gwKEXgM2jfGDxpQnCdosNKY6SGjp_QslDPRj81ZeloPnui45B0_rCgXsgDR16F5NlYs1S1yO1oeDM4Z7GrAjOZFgvWSyABZGkAspeXUyR0wgrukJrYxDoD1qALUVoFJnWpRXdR5SoVzpQInZ2C4HukWc0qt0-okRLwkk5OuRDOlbLUwDPLFYh8KtM2SZbLXJgoOe47XzwWgXokuvDIFB6ZIiLTJqcfU55rvY2_Brc8FF8G1ii0SWcJZhE_yZciQ0uUIYHj-cEv0w7Jmm8mX-fldEgTF9sdkVXztnh4mR-H3fYODzvTQQ
linkProvider	IEEE
linkToHtml	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1Rb9MwED5BQQIeYEDRCmXzw56QvCWxE9uP1dRqFV2FWEF9i1LbJ1WiKdo6fj9nxx1DCCTylAc7ivw55_tyd98BnJTCOFSm4DlmgkvflNxoRF5ho4RqUGHU0vs6U_O5Xi7Np1SsHmthvPcx-cyfhtsYy3dbext-lZ3lMf6oq4fwqJSyyLtyrTt-ZTLRyaNKw2mj6hTEzDNztji_mhIZLIpTQQ6NrsRvx1Dsq_KHMY4nzOTFf77bATxPriQbddi_hAe-fQXP7gkMvgY3atk4VvfxcdSKoGewkbV01sTwOiOXlaVYDZtuyLawwEd_qcqyyfV2w6Yx7Xy9Y0HKg0Z-jumzqWqp7cOXyXhxfsFTXwVuCyN3vMwwQ-JpiKqsmhVROumk8EROXOa8RWfJiWicRpv73JHDqCudS28bAs-tUIo30Gu3rT8EZpVCupRXKyGl941qDIrCCY2yWql8ANl-mWubRMdD74tvdSQfmakDMnVApk7IDODD3ZTvneLGvwb3AxT3BnYoDGC4B7NOH-VNXZAtKojCiertX6Ydw5OLxeWsnk3nH9_B09BavsvSGUKPFt6_h8f2x259c30Ud95PM8DWiA
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=An+Energy-Efficient+Accelerator+for+Medical+Image+Reconstruction+From+Implicit+Neural+Representation&rft.jtitle=IEEE+transactions+on+circuits+and+systems.+I%2C+Regular+papers&rft.au=Rao%2C+Chaolin&rft.au=Wu%2C+Qing&rft.au=Zhou%2C+Pingqiang&rft.au=Yu%2C+Jingyi&rft.date=2023-04-01&rft.pub=The+Institute+of+Electrical+and+Electronics+Engineers%2C+Inc.+%28IEEE%29&rft.issn=1549-8328&rft.eissn=1558-0806&rft.volume=70&rft.issue=4&rft.spage=1625&rft_id=info:doi/10.1109%2FTCSI.2022.3231863&rft.externalDBID=NO_FULL_TEXT
thumbnail_l	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1549-8328&client=summon
thumbnail_m	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1549-8328&client=summon
thumbnail_s	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1549-8328&client=summon