EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Sagittal-Plane Knee Flexion Moment Estimation Using a Lightweight Deep Learning Framework Based on Sequential Surface EMG Feature Frames

Yuanzhi Zhuo iD, Adrian Pranata iD, Chi-Tsun Cheng iD, Toh Yen Pang iD

Sensors · 18.04.2026 · 26 (8) · 2500

DOI: 10.3390/s26082500 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 18.04.2026. Verfasst von Yuanzhi Zhuo und 3 weiteren Autor:innen. Beteiligte Einrichtung: Biomedical Engineering Department, School of Engineering, STEM College, RMIT University, Melbourne, VIC 3000, Australia. Hauptthemen laut MEDLINE: Knee Joint, Electromyography, Deep Learning, Range of Motion, Articular. Die Arbeit nennt 63 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Knee joint moment is an important biomechanical parameter for sports assessment, rehabilitation monitoring, and human–machine interaction. However, direct measurement is often restricted to laboratory-based settings. Surface electromyography (sEMG) offers a non-invasive alternative for indirect joint moment estimation, but many existing deep learning models remain too computationally demanding for potential wearable edge deployment. To address this gap, this study proposes Topo2DCNN-LSTM, a lightweight two-dimensional (2D) convolutional neural network model, designed for sagittal-plane knee flexion moment estimation. The model used a feature-based sequential representation, transforming raw sEMG signals into compact Root Mean Square (RMS) feature frames. The input was processed by a lightweight 2D convolutional neural network (CNN) encoder and paired with long short-term memory (LSTM) units. The model was trained on a public walking dataset of healthy subjects with synchronized sEMG and joint kinetics at two treadmill speeds. When compared with selected deep learning baselines, the quantized model achieved a mean RMS Error of 0.088 ± 0.020 Nm/kg at 1.2 m/s and 0.114 ± 0.034 Nm/kg at 1.8 m/s. On a SparkFun Thing Plus–SAMD51, it achieved an average inference latency of 28 ms using 71,316 bytes of random-access memory (RAM) and 257,172 bytes of flash. These results support its use as a proof of concept for personalized unilateral knee moment estimation with isolated on-device inference feasibility under resource-constrained and limited walking conditions.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Knee JointElectromyographyDeep LearningRange of Motion, ArticularHumansMaleFemaleAdolescentYoung AdultAdultDatasets as TopicWalkingBiomechanical Phenomena

Stichwörter der Autor:innen

continuous motion estimationdeep learningfeature engineeringgait analysisjoint moment estimationknee momentmicrocontrollersurface electromyographywearable sensor

Autor:innen und Institutionen

  • Biomedical Engineering Department, School of Engineering, STEM College, RMIT University, Melbourne, VIC 3000, Australia.
  • School of Health and Biomedical Science, RMIT University, Bundoora, VIC 3082, Australia.
  • Mechanical, Manufacturing and Mechatronic Engineering Department, School of Engineering, STEM College, RMIT University, Melbourne, VIC 3000, Australia.

ORCID-Profile: Yuanzhi Zhuo, Adrian Pranata, Chi-Tsun Cheng, Toh Yen Pang

Zitieren

APA

Zhuo, Y., Pranata, A., Cheng, C. T., & Pang, T. Y. (2026). Sagittal-Plane Knee Flexion Moment Estimation Using a Lightweight Deep Learning Framework Based on Sequential Surface EMG Feature Frames. Sensors, 26(8), 2500. https://doi.org/10.3390/s26082500

Vancouver

Zhuo Y, Pranata A, Cheng CT, Pang TY. Sagittal-Plane Knee Flexion Moment Estimation Using a Lightweight Deep Learning Framework Based on Sequential Surface EMG Feature Frames. Sensors. 2026;26(8):2500. doi:10.3390/s26082500

BibTeX
@article{Zhuo2026,
  author = {Zhuo, Yuanzhi and Pranata, Adrian and Cheng, Chi-Tsun and Pang, Toh Yen},
  title = {Sagittal-Plane Knee Flexion Moment Estimation Using a Lightweight Deep Learning Framework Based on Sequential Surface EMG Feature Frames},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {8},
  pages = {2500},
  doi = {10.3390/s26082500}
}
Kostenlos starten →