EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Energy- and Cost-Efficient Healthcare Task Offloading via Network Edge Digital Twin

Ayesha Jadoon iD, Hao Ran Chi iD, Daniel Corujo iD, Francisco J Ferrão, Rui L Aguiar iD

Sensors · 27.07.2026 · 26 (15) · 4768

DOI: 10.3390/s26154768 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 27.07.2026. Verfasst von Ayesha Jadoon und 4 weiteren Autor:innen. Beteiligte Einrichtung: Instituto de Telecomunicações, University of Aveiro, 3810-193 Aveiro, Portugal. Hauptthemen laut MEDLINE: Delivery of Health Care. Die Arbeit nennt 41 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Science Europe, Fundação para a Ciência e Tecnologia, European Union’s Horizon Europe MSCA-DN programme.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This paper proposes a digital twin (DT)-enabled edge computing framework for healthcare task offloading in smart medical environments. The DT maintains a continuously synchronized virtual representation of the physical edge system, capturing queue states, resource utilization, and energy dynamics. Based on this virtual state, a deterministic optimization model is formulated to support real-time offloading and scheduling decisions under latency, energy, and fairness constraints. Unlike prediction-only approaches, the proposed DT operates in a closed-loop manner, where state estimation and synchronization directly influence scheduling feasibility and system performance. The offloading problem is formulated as a mixed-integer linear programming (MILP) model to jointly optimize task allocation, delay minimization, and energy efficiency. The simulated workload includes 10,000 heterogeneous healthcare applications. At each time slot, one to five tasks are generated and uniformly selected from ECG, video-processing, or medical-imaging workloads, with average task sizes of 90 KB, 280 KB, and 150 KB, respectively. This setup aims to emulate diverse real-world healthcare edge workloads with varying communication and computation demands. The proposed approach significantly reduces average latency by up to 57%, eliminates task drops in all evaluated scenarios, and improves load balancing compared with hospital-only and round-robin baselines. Although total energy consumption increases moderately, energy efficiency per completed task improves due to more effective scheduling by stability, reducing deadline violations and enhancing resource utilization. We further analyze the impact of DT freshness and updated frequency and show that outdated or misaligned DT updates can degrade performance by increasing delay and leading to suboptimal decisions, while overly frequent updates introduce additional coordination overhead. These results highlight the importance of jointly designing DT synchronization mechanisms and optimization-based scheduling strategies for reliable and cost-efficient healthcare edge systems.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Delivery of Health CareHumansAlgorithmsWorkloadCost-Benefit AnalysisDigital Health

Stichwörter der Autor:innen

DT freshnessdigital twinedge computingenergy consumptionhealthcare IoTheuristic approachesresource fairness

Autor:innen und Institutionen

  • Instituto de Telecomunicações, University of Aveiro, 3810-193 Aveiro, Portugal.
  • Data Machine Elite, 2580-654 Lisbon, Portugal.

ORCID-Profile: Ayesha Jadoon, Hao Ran Chi, Daniel Corujo, Rui L Aguiar

Förderung

  • Science Europe
  • Fundação para a Ciência e Tecnologia
  • European Union’s Horizon Europe MSCA-DN programme
  • EU HORIZON JU SNS EXIGENGE
  • FCT/MCTES

Zitieren

APA

Jadoon, A., Chi, H. R., Corujo, D., Ferrão, F. J., & Aguiar, R. L. (2026). Energy- and Cost-Efficient Healthcare Task Offloading via Network Edge Digital Twin. Sensors, 26(15), 4768. https://doi.org/10.3390/s26154768

Vancouver

Jadoon A, Chi HR, Corujo D, Ferrão FJ, Aguiar RL. Energy- and Cost-Efficient Healthcare Task Offloading via Network Edge Digital Twin. Sensors. 2026;26(15):4768. doi:10.3390/s26154768

BibTeX
@article{Jadoon2026,
  author = {Jadoon, Ayesha and Chi, Hao Ran and Corujo, Daniel and Ferrão, Francisco J and Aguiar, Rui L},
  title = {Energy- and Cost-Efficient Healthcare Task Offloading via Network Edge Digital Twin},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {15},
  pages = {4768},
  doi = {10.3390/s26154768}
}
Kostenlos starten →