EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Machine Learning-Driven Optimization of Dielectric Matching Medium for Efficient Wireless Power Transfer in IoMT Implantable Devices

Ashfaq Ahmad, Fawad, Farooq Faisal, Sun-Woong Kim, Dong-You Choi

IEEE Access · 2026 · 14 · 108035-108049

DOI: 10.1109/access.2026.3709745 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Access, 2026. Verfasst von Ashfaq Ahmad und 4 weiteren Autor:innen. Die Arbeit nennt 40 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: research fund from Chosun University, 2026.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This paper presents a machine learning (ML)-assisted optimization framework for enhancing wireless power transfer (WPT) efficiency in Internet of Medical Things (IoMT) implantable systems. The proposed dual-band platform operates at 1.65 GHz for wireless powering and 2.45 GHz for data telemetry, enabling continuous operation and reliable communication within connected healthcare networks. A miniaturized $6 \times 7$ mm implantable receiver (Rx) and a fabrication-friendly slotted-patch transmitter (Tx) are designed to mitigate the challenges posed by lossy biological tissues. To reduce reflection and near-field losses at the tissue interface, an ML-driven dielectric matching medium (MM) is developed using a stacked ensemble model incorporating XGBoost, CatBoost, Decision Tree, and Support Vector Regression (SVR). The optimized MM properties (dielectric constant $\varepsilon _{r} \approx 23$ , and thickness 18 mm) were physically realized using a zirconium dioxide (ZrO2) block and experimentally validated in a tissue-mimicking phantom. Results show a 5 dB improvement in $|S_{21}|$ and an enhancement in WPT efficiency from −22 dB to −17 dB. The study demonstrates the potential of ML-guided electromagnetic design to enable intelligent, energy-efficient, and scalable powering of IoMT implantable devices within future connected healthcare ecosystems.

Förderung

  • research fund from Chosun University, 2026

Zitieren

APA

Ahmad, A., Fawad, Faisal, F., Kim, S., & Choi, D. (2026). Machine Learning-Driven Optimization of Dielectric Matching Medium for Efficient Wireless Power Transfer in IoMT Implantable Devices. IEEE Access, 14, 108035-108049. https://doi.org/10.1109/access.2026.3709745

Vancouver

Ahmad A, Fawad, Faisal F, Kim S, Choi D. Machine Learning-Driven Optimization of Dielectric Matching Medium for Efficient Wireless Power Transfer in IoMT Implantable Devices. IEEE Access. 2026;14:108035-108049. doi:10.1109/access.2026.3709745

BibTeX
@article{Ahmad2026,
  author = {Ahmad, Ashfaq and Fawad and Faisal, Farooq and Kim, Sun-Woong and Choi, Dong-You},
  title = {Machine Learning-Driven Optimization of Dielectric Matching Medium for Efficient Wireless Power Transfer in IoMT Implantable Devices},
  journal = {IEEE Access},
  year = {2026},
  volume = {14},
  pages = {108035--108049},
  doi = {10.1109/access.2026.3709745}
}
Kostenlos starten →