Academic Journal
Προβολή σε EDS

Unified Gait Event Detection using Temporal Convolutional Network and Bayesian Optimization.

Λεπτομέρειες βιβλιογραφικής εγγραφής
Τίτλος: Unified Gait Event Detection using Temporal Convolutional Network and Bayesian Optimization.
Συγγραφείς: Ashraf, Hassan, Schwartz, Cédric, Waris, A, Bruls, Olivier, Boutaayamou, Mohamed
Πηγή: Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 2025, 1 - 4 (2025-07)
Στοιχεία εκδότη: Institute of Electrical and Electronics Engineers Inc., 2025.
Έτος έκδοσης: 2025
Θεματικοί όροι: Humans, Bayes Theorem, Algorithms, Deep Learning, Gait/physiology, Gait Analysis/methods, Neural Networks, Computer, gait event detection, inertial measurement unit, Temporal convolutional network, Bayesian optimization, Convolutional networks, Events detection, Heel strikes, Inertial measurements units, Network optimization, Unified framework, Gait, Gait Analysis, Electrical and Electronic Engineering, Engineering, computing & technology, Ingénierie, informatique & technologie
Περιγραφή: Accurate gait event detection (GED), which involves identifying key events such as heel strikes (HSs) and toe-offs (TOs) from inertial measurement unit (IMU) signals, is critical for quantifying gait abnormalities. However, most existing approaches either rely on heuristic methods tailored to specific activities or require separate deep learning models for different environments, limiting their generalizability and scalability. To address this research gap, we propose a unified deep learning framework based on a temporal convolutional network (TCN) for GED that operates across multiple activities and environments without the need for multiple models. Our approach incorporates: (1) Bayesian optimization for TCN hyperparameter tuning; (2) Gaussian kernel-based ground truth generation; and (3) a weighted loss function to emphasize challenging gait events. Specifically, our unified framework eliminates the need to train separate models for each activity, thereby reducing the computational cost. Evaluation using leave-one-out cross-validation across 20 subjects demonstrates that the proposed framework achieves overall F1 scores of 0.99 ± 0.00 for HS and 0.97 ± 0.06 for TO as well as competitive temporal precision as measured by mean absolute error (MAE). These results outperform existing state-of-the-art methods and underscore the potential of the unified framework for robust GED in diverse real-world scenarios, thereby advancing its application in clinical gait analysis, rehabilitation, and the design of assistive technologies.
Τύπος εγγράφου: journal article
http://purl.org/coar/resource_type/c_6501
article
peer reviewed
Γλώσσα: English
Relation: http://xplorestaging.ieee.org/ielx8/11251507/11251526/11253228.pdf?arnumber=11253228; urn:issn:2694-0604
DOI: 10.1109/EMBC58623.2025.11253228
Σύνδεσμος πρόσβασης: https://orbi.uliege.be/handle/2268/344322
Rights: open access
http://purl.org/coar/access_right/c_abf2
info:eu-repo/semantics/openAccess
Αριθμός Καταχώρησης: edsorb.344322
Βάση Δεδομένων: ORBi
FullText Text:
  Availability: 0
CustomLinks:
  – Url: https://orbi.uliege.be/handle/2268/344322#
    Name: EDS - ORBi (ns324271)
    Category: fullText
    Text: View record at ORBi
Header DbId: edsorb
DbLabel: ORBi
An: edsorb.344322
RelevancyScore: 983
AccessLevel: 3
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 982.763427734375
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Unified Gait Event Detection using Temporal Convolutional Network and Bayesian Optimization.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Ashraf%2C+Hassan%22">Ashraf, Hassan</searchLink><br /><searchLink fieldCode="AR" term="%22Schwartz%2C+Cédric%22">Schwartz, Cédric</searchLink><br /><searchLink fieldCode="AR" term="%22Waris%2C+A%22">Waris, A</searchLink><br /><searchLink fieldCode="AR" term="%22Bruls%2C+Olivier%22">Bruls, Olivier</searchLink><br /><searchLink fieldCode="AR" term="%22Boutaayamou%2C+Mohamed%22">Boutaayamou, Mohamed</searchLink>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 2025, 1 - 4 (2025-07)
– Name: Publisher
  Label: Publisher Information
  Group: PubInfo
  Data: Institute of Electrical and Electronics Engineers Inc., 2025.
– Name: DatePubCY
  Label: Publication Year
  Group: Date
  Data: 2025
– Name: Subject
  Label: Subject Terms
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Humans%22">Humans</searchLink><br /><searchLink fieldCode="DE" term="%22Bayes+Theorem%22">Bayes Theorem</searchLink><br /><searchLink fieldCode="DE" term="%22Algorithms%22">Algorithms</searchLink><br /><searchLink fieldCode="DE" term="%22Deep+Learning%22">Deep Learning</searchLink><br /><searchLink fieldCode="DE" term="%22Gait%2Fphysiology%22">Gait/physiology</searchLink><br /><searchLink fieldCode="DE" term="%22Gait+Analysis%2Fmethods%22">Gait Analysis/methods</searchLink><br /><searchLink fieldCode="DE" term="%22Neural+Networks%2C+Computer%22">Neural Networks, Computer</searchLink><br /><searchLink fieldCode="DE" term="%22gait+event+detection%22">gait event detection</searchLink><br /><searchLink fieldCode="DE" term="%22inertial+measurement+unit%22">inertial measurement unit</searchLink><br /><searchLink fieldCode="DE" term="%22Temporal+convolutional+network%22">Temporal convolutional network</searchLink><br /><searchLink fieldCode="DE" term="%22Bayesian+optimization%22">Bayesian optimization</searchLink><br /><searchLink fieldCode="DE" term="%22Convolutional+networks%22">Convolutional networks</searchLink><br /><searchLink fieldCode="DE" term="%22Events+detection%22">Events detection</searchLink><br /><searchLink fieldCode="DE" term="%22Heel+strikes%22">Heel strikes</searchLink><br /><searchLink fieldCode="DE" term="%22Inertial+measurements+units%22">Inertial measurements units</searchLink><br /><searchLink fieldCode="DE" term="%22Network+optimization%22">Network optimization</searchLink><br /><searchLink fieldCode="DE" term="%22Unified+framework%22">Unified framework</searchLink><br /><searchLink fieldCode="DE" term="%22Gait%22">Gait</searchLink><br /><searchLink fieldCode="DE" term="%22Gait+Analysis%22">Gait Analysis</searchLink><br /><searchLink fieldCode="DE" term="%22Electrical+and+Electronic+Engineering%22">Electrical and Electronic Engineering</searchLink><br /><searchLink fieldCode="DE" term="%22Engineering%2C+computing+%26+technology%22">Engineering, computing & technology</searchLink><br /><searchLink fieldCode="DE" term="%22Ingénierie%2C+informatique+%26+technologie%22">Ingénierie, informatique & technologie</searchLink>
– Name: Abstract
  Label: Description
  Group: Ab
  Data: Accurate gait event detection (GED), which involves identifying key events such as heel strikes (HSs) and toe-offs (TOs) from inertial measurement unit (IMU) signals, is critical for quantifying gait abnormalities. However, most existing approaches either rely on heuristic methods tailored to specific activities or require separate deep learning models for different environments, limiting their generalizability and scalability. To address this research gap, we propose a unified deep learning framework based on a temporal convolutional network (TCN) for GED that operates across multiple activities and environments without the need for multiple models. Our approach incorporates: (1) Bayesian optimization for TCN hyperparameter tuning; (2) Gaussian kernel-based ground truth generation; and (3) a weighted loss function to emphasize challenging gait events. Specifically, our unified framework eliminates the need to train separate models for each activity, thereby reducing the computational cost. Evaluation using leave-one-out cross-validation across 20 subjects demonstrates that the proposed framework achieves overall F1 scores of 0.99 ± 0.00 for HS and 0.97 ± 0.06 for TO as well as competitive temporal precision as measured by mean absolute error (MAE). These results outperform existing state-of-the-art methods and underscore the potential of the unified framework for robust GED in diverse real-world scenarios, thereby advancing its application in clinical gait analysis, rehabilitation, and the design of assistive technologies.
– Name: TypeDocument
  Label: Document Type
  Group: TypDoc
  Data: journal article<br />http://purl.org/coar/resource_type/c_6501<br />article<br />peer reviewed
– Name: Language
  Label: Language
  Group: Lang
  Data: English
– Name: NoteTitleSource
  Label: Relation
  Group: SrcInfo
  Data: http://xplorestaging.ieee.org/ielx8/11251507/11251526/11253228.pdf?arnumber=11253228; urn:issn:2694-0604
– Name: DOI
  Label: DOI
  Group: ID
  Data: 10.1109/EMBC58623.2025.11253228
– Name: URL
  Label: Access URL
  Group: URL
  Data: <link linkTarget="URL" linkTerm="https://orbi.uliege.be/handle/2268/344322" linkWindow="_blank">https://orbi.uliege.be/handle/2268/344322</link>
– Name: Copyright
  Label: Rights
  Group: Cpyrght
  Data: open access<br />http://purl.org/coar/access_right/c_abf2<br />info:eu-repo/semantics/openAccess
– Name: AN
  Label: Accession Number
  Group: ID
  Data: edsorb.344322
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=edsorb&AN=edsorb.344322
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1109/EMBC58623.2025.11253228
    Languages:
      – Text: English
    Subjects:
      – SubjectFull: Humans
        Type: general
      – SubjectFull: Bayes Theorem
        Type: general
      – SubjectFull: Algorithms
        Type: general
      – SubjectFull: Deep Learning
        Type: general
      – SubjectFull: Gait/physiology
        Type: general
      – SubjectFull: Gait Analysis/methods
        Type: general
      – SubjectFull: Neural Networks, Computer
        Type: general
      – SubjectFull: gait event detection
        Type: general
      – SubjectFull: inertial measurement unit
        Type: general
      – SubjectFull: Temporal convolutional network
        Type: general
      – SubjectFull: Bayesian optimization
        Type: general
      – SubjectFull: Convolutional networks
        Type: general
      – SubjectFull: Events detection
        Type: general
      – SubjectFull: Heel strikes
        Type: general
      – SubjectFull: Inertial measurements units
        Type: general
      – SubjectFull: Network optimization
        Type: general
      – SubjectFull: Unified framework
        Type: general
      – SubjectFull: Gait
        Type: general
      – SubjectFull: Gait Analysis
        Type: general
      – SubjectFull: Electrical and Electronic Engineering
        Type: general
      – SubjectFull: Engineering, computing & technology
        Type: general
      – SubjectFull: Ingénierie, informatique & technologie
        Type: general
    Titles:
      – TitleFull: Unified Gait Event Detection using Temporal Convolutional Network and Bayesian Optimization.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Ashraf, Hassan
      – PersonEntity:
          Name:
            NameFull: Schwartz, Cédric
      – PersonEntity:
          Name:
            NameFull: Waris, A
      – PersonEntity:
          Name:
            NameFull: Bruls, Olivier
      – PersonEntity:
          Name:
            NameFull: Boutaayamou, Mohamed
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 07
              Type: published
              Y: 2025
ResultId 1