Individualized Learning-Based Ground Reaction Force Estimation in People Post-Stroke Using Pressure Insoles
Stroke is a leading cause of gait disability that leads to a loss of independence and overall quality of life. The field of clinical biomechanics aims to study how best to provide rehabilitation given an individual's impairments. However, there remains a disconnect between assessment tools used...
Saved in:
| Published in: | IEEE International Conference on Rehabilitation Robotics Vol. 2023; pp. 1 - 6 |
|---|---|
| Main Authors: | , , , , , , , , , |
| Format: | Conference Proceeding Journal Article |
| Language: | English |
| Published: |
IEEE
01.01.2023
|
| ISSN: | 1945-7901, 1945-7901 |
| Online Access: | Get full text |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Abstract | Stroke is a leading cause of gait disability that leads to a loss of independence and overall quality of life. The field of clinical biomechanics aims to study how best to provide rehabilitation given an individual's impairments. However, there remains a disconnect between assessment tools used in biomechanical analysis and in clinics. In particular, 3-dimensional ground reaction forces (3D GRFs) are used to quantify key gait characteristics, but require lab-based equipment, such as force plates. Recent efforts have shown that wearable sensors, such as pressure insoles, can estimate GRFs in real-world environments. However, there is limited understanding of how these methods perform in people post-stroke, where gait is highly heterogeneous. Here, we evaluate three subject-specific machine learning approaches to estimate 3D GRFs with pressure insoles in people post-stroke across varying speeds. We find that a Convolutional Neural Network-based approach achieves the lowest estimation errors of 0.75 ± 0.24, 1.13 ± 0.54, and 4.79 ± 3.04 % bodyweight for the medio-lateral, antero-posterior, and vertical GRF components, respectively. Estimated force components were additionally strongly correlated with the ground truth measurements ( R^{2}> 0.85 ). Finally, we show high estimation accuracy for three clinically relevant point metrics on the paretic limb. These results suggest the potential for an individualized machine learning approach to translate to real-world clinical applications. |
|---|---|
| AbstractList | Stroke is a leading cause of gait disability that leads to a loss of independence and overall quality of life. The field of clinical biomechanics aims to study how best to provide rehabilitation given an individual's impairments. However, there remains a disconnect between assessment tools used in biomechanical analysis and in clinics. In particular, 3-dimensional ground reaction forces (3D GRFs) are used to quantify key gait characteristics, but require lab-based equipment, such as force plates. Recent efforts have shown that wearable sensors, such as pressure insoles, can estimate GRFs in real-world environments. However, there is limited understanding of how these methods perform in people post-stroke, where gait is highly heterogeneous. Here, we evaluate three subject-specific machine learning approaches to estimate 3D GRFs with pressure insoles in people post-stroke across varying speeds. We find that a Convolutional Neural Network-based approach achieves the lowest estimation errors of 0.75 ± 0.24, 1.13 ± 0.54, and 4.79 ± 3.04 % bodyweight for the medio-lateral, antero-posterior, and vertical GRF components, respectively. Estimated force components were additionally strongly correlated with the ground truth measurements ( R^{2}> 0.85 ). Finally, we show high estimation accuracy for three clinically relevant point metrics on the paretic limb. These results suggest the potential for an individualized machine learning approach to translate to real-world clinical applications. Stroke is a leading cause of gait disability that leads to a loss of independence and overall quality of life. The field of clinical biomechanics aims to study how best to provide rehabilitation given an individual's impairments. However, there remains a disconnect between assessment tools used in biomechanical analysis and in clinics. In particular, 3-dimensional ground reaction forces (3D GRFs) are used to quantify key gait characteristics, but require lab-based equipment, such as force plates. Recent efforts have shown that wearable sensors, such as pressure insoles, can estimate GRFs in real-world environments. However, there is limited understanding of how these methods perform in people post-stroke, where gait is highly heterogeneous. Here, we evaluate three subject-specific machine learning approaches to estimate 3D GRFs with pressure insoles in people post-stroke across varying speeds. We find that a Convolutional Neural Network-based approach achieves the lowest estimation errors of 0.75 ± 0.24, 1.13 ± 0.54, and 4.79 ± 3.04 % bodyweight for the medio-lateral, antero-posterior, and vertical GRF components, respectively. Estimated force components were additionally strongly correlated with the ground truth measurements ( ). Finally, we show high estimation accuracy for three clinically relevant point metrics on the paretic limb. These results suggest the potential for an individualized machine learning approach to translate to real-world clinical applications.Stroke is a leading cause of gait disability that leads to a loss of independence and overall quality of life. The field of clinical biomechanics aims to study how best to provide rehabilitation given an individual's impairments. However, there remains a disconnect between assessment tools used in biomechanical analysis and in clinics. In particular, 3-dimensional ground reaction forces (3D GRFs) are used to quantify key gait characteristics, but require lab-based equipment, such as force plates. Recent efforts have shown that wearable sensors, such as pressure insoles, can estimate GRFs in real-world environments. However, there is limited understanding of how these methods perform in people post-stroke, where gait is highly heterogeneous. Here, we evaluate three subject-specific machine learning approaches to estimate 3D GRFs with pressure insoles in people post-stroke across varying speeds. We find that a Convolutional Neural Network-based approach achieves the lowest estimation errors of 0.75 ± 0.24, 1.13 ± 0.54, and 4.79 ± 3.04 % bodyweight for the medio-lateral, antero-posterior, and vertical GRF components, respectively. Estimated force components were additionally strongly correlated with the ground truth measurements ( ). Finally, we show high estimation accuracy for three clinically relevant point metrics on the paretic limb. These results suggest the potential for an individualized machine learning approach to translate to real-world clinical applications. |
| Author | Bergamo, Gregoire Siviy, Christopher Ellis, Terry D. Walsh, Conor J. Novillo, Ignacio Chin, Andrew Wendel, Nicholas Baker, Teresa C. Kim, Daekyum Swaminathan, Krithika |
| Author_xml | – sequence: 1 givenname: Gregoire surname: Bergamo fullname: Bergamo, Gregoire organization: School of Engineering and Applied Sciences, Harvard University,Cambridge,MA – sequence: 2 givenname: Krithika surname: Swaminathan fullname: Swaminathan, Krithika organization: School of Engineering and Applied Sciences, Harvard University,Cambridge,MA – sequence: 3 givenname: Daekyum surname: Kim fullname: Kim, Daekyum organization: School of Engineering and Applied Sciences, Harvard University,Cambridge,MA – sequence: 4 givenname: Andrew surname: Chin fullname: Chin, Andrew organization: School of Engineering and Applied Sciences, Harvard University,Cambridge,MA – sequence: 5 givenname: Christopher surname: Siviy fullname: Siviy, Christopher organization: School of Engineering and Applied Sciences, Harvard University,Cambridge,MA – sequence: 6 givenname: Ignacio surname: Novillo fullname: Novillo, Ignacio organization: School of Engineering and Applied Sciences, Harvard University,Cambridge,MA – sequence: 7 givenname: Teresa C. surname: Baker fullname: Baker, Teresa C. organization: College of Health and Rehabilitation Sciences, Sargent College, Boston University,Boston,MA – sequence: 8 givenname: Nicholas surname: Wendel fullname: Wendel, Nicholas organization: College of Health and Rehabilitation Sciences, Sargent College, Boston University,Boston,MA – sequence: 9 givenname: Terry D. surname: Ellis fullname: Ellis, Terry D. organization: College of Health and Rehabilitation Sciences, Sargent College, Boston University,Boston,MA – sequence: 10 givenname: Conor J. surname: Walsh fullname: Walsh, Conor J. email: walsh@seas.harvard.edu organization: School of Engineering and Applied Sciences, Harvard University,Cambridge,MA |
| BookMark | eNpNkM1OwzAQhA0qEqX0DTj4yCXFP3ESH6GiJVKlVoWeIzveINPULnaCBE9PREHitDuzo0-avUIj5x0ghCmZUUrkXTlfb7eiSJmYMcL4jBJO0kyKMzSVuSy4IDxluSjO0ZjKVCS5JHT0b79E0xjfCCGUFRnLszHal87YD2t61dovMHgFKjjrXpMHFQe5DL53Bm9B1Z31Di98qAE_xs4e1I9hHd6AP7aANz52yXMX_B7wLg4IvAkQYx8Aly76FuI1umhUG2H6Oydot3h8mT8lq_WynN-vEst43iVaQcM1pRo0p7TRyijQWsgclFRGy5obkxndZKlR1FAiJGmGYy1oY4TJgU_Q7Yl7DP69h9hVBxtraFvlwPexYkVREC5Elg3Rm1PUAkB1DEOt8Fn9vZV_AyNEb4o |
| ContentType | Conference Proceeding Journal Article |
| DBID | 6IE 6IL CBEJK RIE RIL 7X8 |
| DOI | 10.1109/ICORR58425.2023.10304695 |
| DatabaseName | IEEE Electronic Library (IEL) Conference Proceedings IEEE Xplore POP ALL IEEE Xplore All Conference Proceedings IEEE/IET Electronic Library (IEL) (UW System Shared) IEEE Proceedings Order Plans (POP All) 1998-Present MEDLINE - Academic |
| DatabaseTitle | MEDLINE - Academic |
| DatabaseTitleList | MEDLINE - Academic |
| Database_xml | – sequence: 1 dbid: RIE name: IEEE Electronic Library (IEL) url: https://ieeexplore.ieee.org/ sourceTypes: Publisher – sequence: 2 dbid: 7X8 name: MEDLINE - Academic url: https://search.proquest.com/medline sourceTypes: Aggregation Database |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Occupational Therapy & Rehabilitation |
| EISBN | 9798350342758 |
| EISSN | 1945-7901 |
| EndPage | 6 |
| ExternalDocumentID | 10304695 |
| Genre | orig-research |
| GrantInformation_xml | – fundername: National Institutes of Health grantid: BRG-ROIHD088619 funderid: 10.13039/100000002 |
| GroupedDBID | 6IE 6IF 6IK 6IL 6IN AAJGR AAWTH ADZIZ ALMA_UNASSIGNED_HOLDINGS BEFXN BFFAM BGNUA BKEBE BPEOZ CBEJK CHZPO IPLJI OCL RIE RIL RNS 7X8 |
| ID | FETCH-LOGICAL-i237t-baef3b11beb311fbadaebb597ea9adb9c3dd6dbf64da1d10590f97ec51fd5d7e3 |
| IEDL.DBID | RIE |
| ISICitedReferencesCount | 4 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001103260000023&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 1945-7901 |
| IngestDate | Fri Jul 11 12:04:38 EDT 2025 Wed Aug 27 02:32:14 EDT 2025 |
| IsPeerReviewed | false |
| IsScholarly | true |
| Language | English |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-i237t-baef3b11beb311fbadaebb597ea9adb9c3dd6dbf64da1d10590f97ec51fd5d7e3 |
| Notes | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23 |
| PQID | 2888035566 |
| PQPubID | 23479 |
| PageCount | 6 |
| ParticipantIDs | proquest_miscellaneous_2888035566 ieee_primary_10304695 |
| PublicationCentury | 2000 |
| PublicationDate | 2023-01-01 |
| PublicationDateYYYYMMDD | 2023-01-01 |
| PublicationDate_xml | – month: 01 year: 2023 text: 2023-01-01 day: 01 |
| PublicationDecade | 2020 |
| PublicationTitle | IEEE International Conference on Rehabilitation Robotics |
| PublicationTitleAbbrev | ICORR |
| PublicationYear | 2023 |
| Publisher | IEEE |
| Publisher_xml | – name: IEEE |
| SSID | ssj0001286276 |
| Score | 2.2363915 |
| Snippet | Stroke is a leading cause of gait disability that leads to a loss of independence and overall quality of life. The field of clinical biomechanics aims to study... |
| SourceID | proquest ieee |
| SourceType | Aggregation Database Publisher |
| StartPage | 1 |
| Title | Individualized Learning-Based Ground Reaction Force Estimation in People Post-Stroke Using Pressure Insoles |
| URI | https://ieeexplore.ieee.org/document/10304695 https://www.proquest.com/docview/2888035566 |
| Volume | 2023 |
| WOSCitedRecordID | wos001103260000023&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/eLvHCXMwlV1LS8NAEF5s8eDJV8X6KCuot23zaHaTq6XFXmqpFXoL-4qEQiJp6sFf7-wmqYp48BZIhg27w-y8vvkQug3DJAiZZoSBb0-G4GGTyHcFgQAokEEgBY8SSzbBZrNwtYrmNVjdYmG01rb5TPfNo63lq1xuTaps4No6XhS0UIsxWoG1viVUwDlntOnWcaLBdPS0WASmztQ3JOH9RrwmUvllfe2VMjn8588coc4XOA_Pd9fOMdrT2Qm6-z4tGC-rUQH4Hi9-DOI-RevpDoCVfmiF6_Gqr-QBbjOFTSYqUyBVwR3wJC9gtTGYgQrhiNMMz23POTYsv-S5LPK1xrbvAFdIw0LjabYxY6I66GUyXo4eSc23QFLPZyURXCe-cF0BAbbrJoIrroWAiEPziCsRSd_STyV0qLirjGPmJPBSBm6iAsW0f4baWZ7pc4SdSEiIHKUCQzwUvuQ0pPAJ8yg4GI7Hu6hjdjR-q0ZqxM1mdtFNcyYx6LkpXvBM59tN7EGo7oBzROnFH7KX6MAcc5UfuULtstjqa7Qv38t0U_RAZVZhz6rMJ9hVw_I |
| linkProvider | IEEE |
| linkToHtml | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1JS8NAGB20Cnpyq1jXEdTb1GyT5WppabDWUiv0FmaLlEIiaevBX-83k7RWxIO3QDJJmBm--bb3HkI3YZjSMFABCcC3Jx542CRybU4gAKKCUsFZlBqxiaDfD8fjaFCB1Q0WRillms9UU1-aWr7MxUKnyu5tU8eL6Cbaop7nWCVcay2lAu554C_7dazoPm49D4dUV5qaWia8uXxBJaXyy_6aQ6Wz98_f2Uf1b3geHqwOngO0obJDdLvOF4xHJVkAvsPDH1TcR2garyBYk08lcUWw-kYe4DyTWOeiMgmjSsAD7uQFfK0NhqDEOOJJhgem6xxrnV_yMi_yqcKm8wCXWMNC4TibaaKoOnrttEetLqkUF8jEcYM54UylLrdtDiG2baecSaY4h5hDsYhJHgnXCFClvieZLbVrZqVwU1A7lVQGyj1GtSzP1AnCVsQFxI5Cgin2uCuYH_rwSOD44GJYDmugup7R5L0k1UiWk9lA18s1SWCn6_IFy1S-mCUOBOsWuEe-f_rH2Cu00x099ZJe3H88Q7t6yctsyTmqzYuFukDb4mM-mRWXZuN8AY0VxlE |
| 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%3Abook&rft.genre=proceeding&rft.title=IEEE+International+Conference+on+Rehabilitation+Robotics&rft.atitle=Individualized+Learning-Based+Ground+Reaction+Force+Estimation+in+People+Post-Stroke+Using+Pressure+Insoles&rft.au=Bergamo%2C+Gregoire&rft.au=Swaminathan%2C+Krithika&rft.au=Kim%2C+Daekyum&rft.au=Chin%2C+Andrew&rft.date=2023-01-01&rft.pub=IEEE&rft.eissn=1945-7901&rft.spage=1&rft.epage=6&rft_id=info:doi/10.1109%2FICORR58425.2023.10304695&rft.externalDocID=10304695 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1945-7901&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1945-7901&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1945-7901&client=summon |