Abstract
Clinical large language model (LLM) agents deployed locally can consult more capable remote models, but doing so risks exposing patient information. Privacy-conscious delegation places disclosure decisions with a local agent, yet removing explicit identifiers is insufficient: quasi-identifiers can accumulate across multi-turn consultations and repeated patient visits to enable re-identification. We introduce PrivMeSA, a privacy-aware self-evolving multi-agent system that learns to control disclosure and retains remote expertise for local reuse. A local agent manages each encounter and consults remote specialists that may request additional information. Reinforcement learning balances task accuracy against direct disclosure and registry-based re-identification risk, with privacy evaluated over the complete outbound transcript of each encounter. A local lesson memory distills completed consultations into generalized clinical guidance and retrieves relevant lessons before transmission, allowing subsequent cases to reuse expertise without another remote exchange. Memory grows without additional outcome labels or parameter updates. On an emergency-department benchmark built from MIMIC-IV-ED records, PrivMeSA improves mean task accuracy over delegation by up to 15.8 percentage points. In the same setting, PrivMeSA reduces the disclosure of personal details from 98.0% to 0.2% of cases and the share of cases in which the patient can be narrowed to ten or fewer registry patients from 74% to 0%.
Keywords
Subject
Publication details
- Journal
- Not available
- Open access
- Green open access
Cite this article
APA 7
Wang, D., Zhang, Y., Sun, B., Stinard, A., Shang, Y., Wang, S., & Tian, Y. (2026). PrivMeSA: Privacy-Aware Self-Evolving Multi-Agent System for Medicine via Local-Remote LLM Collaboration. https://omanscience.com/en/articles/privmesa-privacy-aware-self-evolving-multi-agent-system-for-medicine-via-local-remote-llm-collaboration
MLA 9
Wang, Dannong, et al. "PrivMeSA: Privacy-Aware Self-Evolving Multi-Agent System for Medicine via Local-Remote LLM Collaboration." https://omanscience.com/en/articles/privmesa-privacy-aware-self-evolving-multi-agent-system-for-medicine-via-local-remote-llm-collaboration.
Chicago (author–date)
Wang, Dannong, Yuran Zhang, Bian Sun, Alex Stinard, Yuzhang Shang, Song Wang, and Yu Tian. 2026. "PrivMeSA: Privacy-Aware Self-Evolving Multi-Agent System for Medicine via Local-Remote LLM Collaboration." https://omanscience.com/en/articles/privmesa-privacy-aware-self-evolving-multi-agent-system-for-medicine-via-local-remote-llm-collaboration.
Harvard
Wang, D., Zhang, Y., Sun, B., Stinard, A., Shang, Y., Wang, S. and Tian, Y. (2026) 'PrivMeSA: Privacy-Aware Self-Evolving Multi-Agent System for Medicine via Local-Remote LLM Collaboration', Available at: https://omanscience.com/en/articles/privmesa-privacy-aware-self-evolving-multi-agent-system-for-medicine-via-local-remote-llm-collaboration.
Vancouver
Wang D, Zhang Y, Sun B, Stinard A, Shang Y, Wang S, et al. PrivMeSA: Privacy-Aware Self-Evolving Multi-Agent System for Medicine via Local-Remote LLM Collaboration. https://omanscience.com/en/articles/privmesa-privacy-aware-self-evolving-multi-agent-system-for-medicine-via-local-remote-llm-collaboration
IEEE
D. Wang, Y. Zhang, B. Sun, A. Stinard, Y. Shang, S. Wang, and Y. Tian, "PrivMeSA: Privacy-Aware Self-Evolving Multi-Agent System for Medicine via Local-Remote LLM Collaboration," https://omanscience.com/en/articles/privmesa-privacy-aware-self-evolving-multi-agent-system-for-medicine-via-local-remote-llm-collaboration.