Privacy-Preserving LLM Agent for Multi-modal Health Monitoring

  • Qipeng Xie
  • , Jiafei Wu
  • , Weiyu Wang
  • , Zhuotao Lian
  • , Mu Yuan
  • , Xian Shuai
  • , Weizheng Wang
  • , Yuan Haoyi
  • , Haibo Hu
  • , Kaishun Wu

Research output: Chapter in book / Conference proceedingConference article published in proceeding or bookAcademic researchpeer-review

Abstract

Tool-using LLM agents for health monitoring raise critical privacy concerns as they share sensitive patient data with cloud providers and third-party models. This study presents HealthAgent, a privacy-preserving LLM agent framework that protects both user queries and multi-modal sensor data through homomorphic encryption. HealthAgent enables an LLM orchestrator to coordinate specialized AI models for complex health assessments while processing all data in encrypted form. The system achieves 95% task decomposition accuracy with 10 s latency, demonstrating that strong privacy guarantees can be maintained without sacrificing real-time performance in health monitoring applications.

Original languageEnglish
Title of host publicationProvable and Practical Security - 19th International Conference, ProvSec 2025, Proceedings
EditorsGuomin Yang, Shengli Liu, Chunhua Su, Akira Otsuka, Zhuotao Lian
PublisherSpringer Science and Business Media Deutschland GmbH
Pages488-492
Number of pages5
ISBN (Print)9789819529605
DOIs
Publication statusPublished - Nov 2025
Event19th International Conference on Provable and Practical Security, ProvSec 2025 - Yokohama, Japan
Duration: 10 Oct 202512 Oct 2025

Publication series

NameLecture Notes in Computer Science
Volume16172 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference19th International Conference on Provable and Practical Security, ProvSec 2025
Country/TerritoryJapan
CityYokohama
Period10/10/2512/10/25

Keywords

  • Health
  • Homomorphic encryption
  • Large language models (LLM)
  • Privacy

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science

Fingerprint

Dive into the research topics of 'Privacy-Preserving LLM Agent for Multi-modal Health Monitoring'. Together they form a unique fingerprint.

Cite this