EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Privacy‐preserving multi‐agent systems for human‐centred inclusive healthcare AI

Mufti Mahmud, Noushath Shaffi, M Shamim Kaiser, M Arifur Rahman, M Mostafizur Rahman, Vimbi Viswan, Tamanna Sharmeen, Sweta Bhattacharya, Neelu Jyoti Ahuja, Ayse Ulgen, Imran Khan Niazi, Kanad Ray und 3 weitere

AI Magazine · 2026-09 · 47 (3)

DOI: 10.1002/aaai.70087 ↗

Kurz gefasst

Fachartikel in der Zeitschrift AI Magazine, online erschienen am 24.08.2026. Verfasst von Mufti Mahmud und 14 weiteren Autor:innen. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 20 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: King Fahd University of Petroleum and Minerals, Nottingham Trent University.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Healthcare environments present uniquely demanding constraints for the deployment of artificial intelligence (AI). Clinical decisions of significant consequence are rarely the product of a single isolated computation; rather, safe and effective patient care relies on the continuous collaboration of distributed entities, including clinicians, hospital networks, laboratory systems, monitoring devices, and patients themselves. As generative models and large language models are increasingly deployed in clinical settings, the transition from monolithic AI models to Multi‐Agent Systems (MAS) offers unprecedented capabilities for distributed monitoring, coordinated decision support, and service integration. However, multi‐agent collaboration simultaneously introduces severe novel vulnerabilities, including topological contamination, privacy leakage, cascading systemic failures, and the socio‐psychological risks of undue algorithmic persuasion. This work introduces a comprehensive, privacy‐preserving, explainable, and human‐centred MAS framework explicitly designed for collaborative clinical decision support. Rejecting the paradigm of a single, omniscient AI model with unrestricted data access, the proposed architecture operationalises clinical AI as a heavily regulated team of specialised, interacting agents. The framework defines six distinct agent classes: Data Stewardship, Modality, Explanation, Consensus, Safety Monitor, and Clinician Interface agents. Furthermore, the implementation is governed by six foundational design principles grounded in classic Autonomous Agents and MAS theory, adapted for modern generative systems. By enforcing minimum‐necessary data exchange, verifiable team‐level explainability, explicit hierarchical escalation, secure communication topologies, and human‐centred safety protocols, this framework provides both a domain‐grounded blueprint and a transferable set of robust design principles for deploying trustworthy MAS in high‐stakes environments.

Förderung

  • King Fahd University of Petroleum and Minerals
  • Nottingham Trent University

Zitieren

APA

Mahmud, M., Shaffi, N., Kaiser, M. S., Rahman, M. A., Rahman, M. M., Viswan, V., Sharmeen, T., Bhattacharya, S., Ahuja, N. J., Ulgen, A., Niazi, I. K., Ray, K., Hussain, A., Brown, D. J., & Moore, J. H. (2026). Privacy‐preserving multi‐agent systems for human‐centred inclusive healthcare AI. AI Magazine, 47(3). https://doi.org/10.1002/aaai.70087

Vancouver

Mahmud M, Shaffi N, Kaiser MS, Rahman MA, Rahman MM, Viswan V, et al. Privacy‐preserving multi‐agent systems for human‐centred inclusive healthcare AI. AI Magazine. 2026;47(3). doi:10.1002/aaai.70087

BibTeX
@article{Mahmud2026,
  author = {Mahmud, Mufti and Shaffi, Noushath and Kaiser, M Shamim and Rahman, M Arifur and Rahman, M Mostafizur and Viswan, Vimbi and Sharmeen, Tamanna and Bhattacharya, Sweta and Ahuja, Neelu Jyoti and Ulgen, Ayse and Niazi, Imran Khan and Ray, Kanad and Hussain, Amir and Brown, David J and Moore, Jason H.},
  title = {Privacy‐preserving multi‐agent systems for human‐centred inclusive healthcare AI},
  journal = {AI Magazine},
  year = {2026},
  volume = {47},
  number = {3},
  doi = {10.1002/aaai.70087}
}
Kostenlos starten →