EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Deep Learning 1D-CNN-Based Ground Contact Detection in Sprint Acceleration Using Inertial Measurement Units

Felix Friedl iD, Thorben Menrad, Jürgen Edelmann-Nusser

Sensors · 05.01.2026 · 26 (1) · 342

DOI: 10.3390/s26010342 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 05.01.2026. Verfasst von Felix Friedl und 2 weiteren Autor:innen. Beteiligte Einrichtung: Sports and Technology, Institute of Sport Science, Otto von Guericke University Magdeburg, 39116 Magdeburg, Germany. Hauptthemen laut MEDLINE: Deep Learning, Running. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 40 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Federal Ministry of the Interior, AiF Projekt GmbH.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Ground contact (GC) detection is essential for sprint performance analysis. Inertial measurement units (IMUs) enable field-based assessment, but their reliability during sprint acceleration remains limited when using heuristic and recently used machine learning algorithms. This study introduces a deep learning one-dimensional convolutional neural network (1D-CNN) to improve GC event and GC times detection in sprint acceleration.

Methods

Twelve sprint-trained athletes performed 60 m sprints while bilateral shank-mounted IMUs (1125 Hz) and synchronized high-speed video (250 Hz) captured the first 15 m. Video-derived GC events served as reference labels for model training, validation, and testing, using resultant acceleration and angular velocity as model inputs.

Results

The optimized model (18 inception blocks, window = 100, stride = 15) achieved mean Hausdorff distances ≤ 6 ms and 100% precision and recall for both validation and test datasets (Rand Index ≥ 0.977). Agreement with video references was excellent (bias < 1 ms, limits of agreement ± 15 ms, r > 0.90, p < 0.001).

Conclusions

The 1D-CNN surpassed heuristic and prior machine learning approaches in the sprint acceleration phase, offering robust, near-perfect GC detection. These findings highlight the promise of deep learning-based time-series models for reliable, real-world biomechanical monitoring in sprint acceleration tasks.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningRunningHumansConvolutional Neural NetworksAccelerationAlgorithmsAthletic PerformanceMaleBiomechanical Phenomena

Stichwörter der Autor:innen

IMUInertial measurement unitsbiomechanicsconvolutional neural networkdeep learningfield-based monitoringground contact detectionmotion analysisspatiotemporalsports performance analysissprint accelerationtime-series classification

Autor:innen und Institutionen

  • Sports and Technology, Institute of Sport Science, Otto von Guericke University Magdeburg, 39116 Magdeburg, Germany.
  • Department Team, Combat & Acrobatic Sports, Institute for Applied Training Science, 04109 Leipzig, Germany.

ORCID-Profile: Felix Friedl

Förderung

  • Federal Ministry of the Interior
  • AiF Projekt GmbH

Zitieren

APA

Friedl, F., Menrad, T., & Edelmann-Nusser, J. (2026). Deep Learning 1D-CNN-Based Ground Contact Detection in Sprint Acceleration Using Inertial Measurement Units. Sensors, 26(1), 342. https://doi.org/10.3390/s26010342

Vancouver

Friedl F, Menrad T, Edelmann-Nusser J. Deep Learning 1D-CNN-Based Ground Contact Detection in Sprint Acceleration Using Inertial Measurement Units. Sensors. 2026;26(1):342. doi:10.3390/s26010342

BibTeX
@article{Friedl2026,
  author = {Friedl, Felix and Menrad, Thorben and Edelmann-Nusser, Jürgen},
  title = {Deep Learning 1D-CNN-Based Ground Contact Detection in Sprint Acceleration Using Inertial Measurement Units},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {1},
  pages = {342},
  doi = {10.3390/s26010342}
}
Kostenlos starten →