EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Explainable AI for gait speed analysis from multimodal data fusion

Abdullah Alharthi iD, Abdulrahman Al Ayidh, Ahmed Alqurashi, Mohammed Alammar, Shoaib Shahriar Mohammad, Houssem Bouchekara iD, Yusuf Abubakar Sha'aban, Turki Essa Alharbi

PLOS One · 2026

DOI: 10.1371/journal.pone.0341067 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS One, 2026. Verfasst von Abdullah Alharthi und 7 weiteren Autor:innen. Beteiligte Einrichtung: Department of Electrical Engineering, King Khalid University, Abha, Saudi Arabia. Hauptthemen laut MEDLINE: Walking Speed, Gait, Gait Analysis. Die Arbeit nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Deanship of Scientific Research, King Khalid University.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Gait speed analysis is vital for applications in healthcare, rehabilitation, human-robot interaction, and autonomous systems, necessitating robust methods to address its complexity. This study introduces an advanced framework for gait speed classification through multimodal data fusion and deep learning, optimized using Layer-wise Relevance Propagation (LRP). Utilizing a publicly available 4 datasets of 50 injury-free adults walking at varying speeds, we integrate data from full-body motion capture, electromyography (EMG), and force plates to capture detailed gait dynamics. We propose two novel architectures: a hybrid Convolutional Neural Network with Long Short-Term Memory (CNN+LSTM) and a Multi-stream CNN, benchmarked against Temporal Convolutional Networks (TCNs), Transformer neural networks, Gated Recurrent Units (GRU) and statistical classifiers (e.g., Linear Discriminant Analysis, Quadratic Discriminant Analysis, Support Vector Machines). All models in this study were trained, validated and tested using three distinct strategies: experiment 1 (sample-based splitting – 10-fold cross-validation), experiment 2 (10-fold cross-validation, subject-based splitting – 5 subjects for testing), and experiment 3 (5-fold cross-validation, subject-based splitting – 10 subjects for testing). Multi-stream Quads CNN is the best-performing model, achieving the highest F1 scores across all experiments (96.6% ± 1.4% experiment 1, 96.2% ± 1.6% experiment 2, 95.9% ± 0.9% experiment 3 with top performance for all experiments 98% ± 0.6%, outperforming traditional approaches. LRP is applied to evaluate feature relevance within the model, enabling the removal of critical features to verify robustness. Comprehensive assessments, including ROC curves, confusion matrices, t-test, and perturbation analyses, validate the model’s enhanced performance and durability. By combining multimodal fusion to overcome single-sensor constraints with LRP-driven feature optimization, this work provides a highly accurate and resilient tool for gait analysis, with significant potentials.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Walking SpeedGaitGait AnalysisHumansConvolutional Neural NetworksLong Short Term MemoryElectromyographyAdultSupport Vector MachineNeural Networks, ComputerMaleDeep LearningDiscriminant AnalysisMotion CaptureFemale

Autor:innen und Institutionen

  • Department of Electrical Engineering, King Khalid University, Abha, Saudi Arabia.
  • Department of Electrical Engineering, Umm Al Qura University, Makkah, Saudi Arabia.
  • Department of Electrical Engineering, University of Hafr Al Batin, Hafr Al Batin, Saudi Arabia.
  • Department of Computer and Software Engineering, Embry-Riddle Aeronautical University, Prescott, Arizona, United States of America.
  • Center for International Studies, Massachusetts Institute of Technology, Cambridge, Massachusetts, United States of America,.
  • Department of Electrical Engineering, College of Engineering, Taif University, Taif, Saudi Arabia.

ORCID-Profile: Abdullah Alharthi, Houssem Bouchekara

Förderung

  • Deanship of Scientific Research, King Khalid University

Zitieren

APA

Alharthi, A., Al Ayidh, A., Alqurashi, A., Alammar, M., Mohammad, S. S., Bouchekara, H., Sha'aban, Y. A., & Alharbi, T. E. (2026). Explainable AI for gait speed analysis from multimodal data fusion. PLOS One. https://doi.org/10.1371/journal.pone.0341067

Vancouver

Alharthi A, Al Ayidh A, Alqurashi A, Alammar M, Mohammad SS, Bouchekara H, et al. Explainable AI for gait speed analysis from multimodal data fusion. PLOS One. 2026. doi:10.1371/journal.pone.0341067

BibTeX
@article{Alharthi2026,
  author = {Alharthi, Abdullah and Al Ayidh, Abdulrahman and Alqurashi, Ahmed and Alammar, Mohammed and Mohammad, Shoaib Shahriar and Bouchekara, Houssem and Sha'aban, Yusuf Abubakar and Alharbi, Turki Essa},
  title = {Explainable AI for gait speed analysis from multimodal data fusion},
  journal = {PLOS One},
  year = {2026},
  doi = {10.1371/journal.pone.0341067}
}
Kostenlos starten →