EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Lightweight Deep Learning Approaches on Edge Devices for Fetal Movement Monitoring

Atcharawan Rattanasak iD, Talit Jumphoo, Kasidit Kokkhunthod iD, Wongsathon Pathonsuwan, Rattikan Nualsri, Sittinon Thanonklang, Pattama Tongdee, Porntip Nimkuntod, Monthippa Uthansakul, Peerapong Uthansakul iD

Biosensors · 2025

DOI: 10.3390/bios15100662 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Biosensors, 2025. Verfasst von Atcharawan Rattanasak und 9 weiteren Autor:innen. Beteiligte Einrichtung: School of Telecommunication Engineering, Suranaree University of Technology, Nakhon Ratchasima 30000, Thailand. Hauptthemen laut MEDLINE: Deep Learning, Fetal Movement, Fetal Monitoring. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 44 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Suranaree University of Technology (SUT), Thailand Science Research and Innovation (TSRI) and National Science, Research and Innovation Fund (NSRF) through NRI…, National Research Council of Thailand (NRCT).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Fetal movement monitoring (FMM) is crucial for assessing fetal well-being, traditionally relying on clinical assessments or maternal perception, each with inherent limitations. This study presents a novel lightweight deep learning framework for real-time FMM on edge devices. Data were collected from 120 participants using a wearable device equipped with an inertial measurement unit, which captured both accelerometer and gyroscope data, coupled with a rigorous two-stage labeling protocol integrating maternal perception and ultrasound validation. We addressed class imbalance using virtual-rotation-based augmentation and adaptive clustering-based undersampling. The data were transformed into spectrograms using the Short-Time Fourier Transform, serving as input for deep learning models. To ensure model efficiency suitable for resource-constrained microcontrollers, we employed knowledge distillation, transferring knowledge from larger, high-performing teacher models to compact student architectures. Post-training integer quantization further optimized the models, reducing the memory footprint by 74.8%. The final optimized model achieved a sensitivity (SEN) of 90.05%, a precision (PRE) of 87.29%, and an F1-score (F1) of 88.64%. Practical energy assessments showed continuous operation capability for approximately 25 h on a single battery charge. Our approach offers a practical framework adaptable to other medical monitoring tasks on edge devices, paving the way for improved prenatal care, especially in resource-limited settings.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningFetal MovementFetal MonitoringHumansFemalePregnancyWearable Electronic Devices

Stichwörter der Autor:innen

deep learningembedded systemfetal movement detectionknowledge distillation

Autor:innen und Institutionen

  • School of Telecommunication Engineering, Suranaree University of Technology, Nakhon Ratchasima 30000, Thailand.
  • Institute of Research and Development, Suranaree University of Technology, Nakhon Ratchasima 30000, Thailand.
  • School of Obstetrics and Gynecology, Institute of Medicine, Suranaree University of Technology, Nakhon Ratchasima 30000, Thailand.
  • School of Medicine, Institute of Medicine, Suranaree University of Technology, Nakhon Ratchasima 30000, Thailand.

ORCID-Profile: Atcharawan Rattanasak, Kasidit Kokkhunthod, Peerapong Uthansakul

Förderung

  • Suranaree University of Technology (SUT), Thailand Science Research and Innovation (TSRI) and National Science, Research and Innovation Fund (NSRF) through NRI…
  • National Research Council of Thailand (NRCT)

Zitieren

APA

Rattanasak, A., Jumphoo, T., Kokkhunthod, K., Pathonsuwan, W., Nualsri, R., Thanonklang, S., Tongdee, P., Nimkuntod, P., Uthansakul, M., & Uthansakul, P. (2025). Lightweight Deep Learning Approaches on Edge Devices for Fetal Movement Monitoring. Biosensors. https://doi.org/10.3390/bios15100662

Vancouver

Rattanasak A, Jumphoo T, Kokkhunthod K, Pathonsuwan W, Nualsri R, Thanonklang S, et al. Lightweight Deep Learning Approaches on Edge Devices for Fetal Movement Monitoring. Biosensors. 2025. doi:10.3390/bios15100662

BibTeX
@article{Rattanasak2025,
  author = {Rattanasak, Atcharawan and Jumphoo, Talit and Kokkhunthod, Kasidit and Pathonsuwan, Wongsathon and Nualsri, Rattikan and Thanonklang, Sittinon and Tongdee, Pattama and Nimkuntod, Porntip and Uthansakul, Monthippa and Uthansakul, Peerapong},
  title = {Lightweight Deep Learning Approaches on Edge Devices for Fetal Movement Monitoring},
  journal = {Biosensors},
  year = {2025},
  doi = {10.3390/bios15100662}
}
Kostenlos starten →