Adaptive Edge-to-Cloud AI Inference Frameworks for Low-Latency Decision Support
Keywords:
Edge computing, cloud computing, artificial intelligence, AI inferenceAbstract
The increasing deployment of artificial intelligence (AI) in healthcare and other time-sensitive decision environments has created a need for inference architectures that simultaneously provide low latency, scalability, privacy, interpretability, and operational resilience. Conventional cloud-centric AI pipelines offer substantial computational capacity but may introduce communication delays, bandwidth dependence, privacy exposure, and reduced responsiveness when large volumes of heterogeneous data must be processed continuously. This research and review paper examines adaptive edge-to-cloud AI inference frameworks as an architectural approach for addressing these limitations. The study synthesizes the provided literature on healthcare big-data management, multimodal biomedical AI, explainable AI, federated learning, privacy-preserving architectures, collaborative neural networks, blockchain-enabled medical analysis, and secure access control. Based on this synthesis, an adaptive framework is conceptually developed around workload-aware inference placement, dynamic model selection, privacy-preserving distributed learning, explainability, and resilient edge-to-cloud orchestration. The analysis indicates that effective edge-to-cloud inference should not be understood merely as moving computation closer to data; rather, it requires coordinated decisions concerning where inference occurs, which model is deployed, what information is transmitted, and how privacy and interpretability requirements influence execution. The proposed framework provides a theoretical foundation for low-latency decision support while recognizing trade-offs involving computational constraints, model complexity, communication overhead, security, and consistency. The findings further indicate that adaptive orchestration is particularly important for healthcare applications in which real-time responsiveness must coexist with stringent privacy and clinical interpretability requirements.
Downloads
References
1. K. Abouelmehdi, “Big data security and privacy in healthcare: A review,” Procedia Comput. Sci., vol. 113, pp. 73–80, 2017.
2. J. N. Acosta, “Multimodal biomedical AI,” Nature Med., vol. 28, no. 9, pp. 1773–1784, 2022.
3. I. Ahmad, M. Zhu, G. Li, D. Javeed, P. Kumar, and S. Chen, “A secure and interpretable AI for smart healthcare system: A case study on epilepsy diagnosis using EEG signals,” IEEE J. Biomed. Health Inform., early access, Mar. 20, 2024, doi: 10.1109/JBHI.2024.3366341.
4. S. Ali, “Explainable artificial intelligence (XAI): What we know and what is left to attain trustworthy artificial intelligence,” Inf. Fusion, vol. 99, 2023, Art. no. 101805.
5. S. Bharati, M. R. H. Mondal, P. Podder, and U. Kose, “Explainable artificial intelligence (XAI) with IoHT for smart healthcare: A review,” in Interpretable Cognitive Internet of Things for Healthcare. Berlin, Germany : Springer, 2023, pp. 1–24.
6. Z. Chen, J. Du, X. Hou, K. Yu, J. Wang, and Z. Han, “Channel adaptive and sparsity personalized federated learning for privacy protection in smart healthcare systems,” IEEE J. Biomed. Health Inform., early access, Jan. 15, 2024, doi: 10.1109/JBHI.2024.3353791.
7. S. Dash, “Big data in healthcare: Management, analysis and future prospects,” J. Big Data, vol. 6, no. 1, pp. 1–25, 2019.
8. S. Devi, M. N. Sahoo, and S. Bakshi, “Explainable reverse verification of goodness of classification of MRI images by clinical experts,” IEEE J. Biomed. Health Inform., early access, May 26, 2023, doi: 10.1109/JBHI.2023.3280184.
9. I. A. Khan, I. Razzak, D. Pi, U. Zia, S. Kamal, and Y. Hussain, “A novel collaborative SRU network with dynamic behaviour aggregation, reduced communication overhead and explainable features,” IEEE J. Biomed. Health Inform., early access, Jan. 10, 2024, doi: 10.1109/JBHI.2024.3352013.
10. X. Liang, Y. Liu, and J. Ning, “An access control scheme with privacy-preserving authentication and flexible revocation for smart healthcare,” IEEE J. Biomed. Health Inform., early access, Apr. 24, 2024, doi: 10.1109/JBHI.2024.3391218.
11. G. Manogaran, “A new architecture of internet of Things and Big Data ecosystem for secured smart healthcare monitoring and alerting system,” Future Gener. Comput. Syst., vol. 82 : pp. 375–387, 2018.
12. J. Mu, M. Kadoch, T. Yuan, W. Lv, Q. Liu, and B. Li, “Explainable federated medical image analysis through causal learning and blockchain,” IEEE J. Biomed. Health Inform., early access, Mar. 18, 2024, doi: 10.1109/JBHI.2024.3375894.
13. D. Saraswat, “Explainable AI for healthcare 5.0: Opportunities and challenges,” IEEE Access, vol. 10, pp. 84486–84517, 2022.
14. Z. Tang, H. S. Wong, and Z. Yu, “Privacy-preserving federated learning with domain adaptation for multi-disease ocular disease recognition,” IEEE J. Biomed. Health Inform., early access, Aug. 17, 2023, doi: 10.1109/JBHI.2023.3305685.
15. K. Kumar, "Edge-to-Cloud AI Inference Pipelines: A Resilient Architecture for Real-Time Decision Systems," 2025 International Conference on Computer and Applications (ICCA), Bahrain, Bahrain, 2025, pp. 1-6, doi: 10.1109/ICCA66035.2025.11430905.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Dr. Mustafa Al-Sadi, Dr. Rana Al-Jubouri (Author)
This work is licensed under a Creative Commons Attribution 4.0 International License.
All articles published by The Parthenon Frontiers and its associated journals are distributed under the terms of the Creative Commons Attribution (CC BY 4.0) International License unless otherwise stated.
Authors retain full copyright of their published work. By submitting their manuscript, authors agree to grant The Parthenon Frontiers a non-exclusive license to publish, archive, and distribute the article worldwide. Authors are free to:
-
Share their article on personal websites, institutional repositories, or social media platforms.
-
Reuse their content in future works, presentations, or educational materials, provided proper citation of the original publication.
