Yung Wei Shueh, Zhi-Jie Chen, Chia-Hsuan Hsu, Hsin-Ling Hsu, Donghua Zhang, Chenwei Wu, Jun-En Ding, Tongze Zhang, Shihao Yang, Pengfei Hu, Fang-Ming Hung, Feng Liu
DIASENTINEL is a multi-agent system for type 2 diabetes risk screening and report generation from electronic health records. It integrates risk prediction, clinical signal extraction, and verification layers for reliable clinical decision support. The system provides a practical framework for auditable and privacy-preserving large language model-based decision support.
DIASENTINEL addresses the vulnerabilities of large language models in clinical decision support, such as hallucinated facts and unsupported recommendations. The system's auditable and privacy-preserving approach can increase trust in AI-based clinical decision support. This can lead to better patient outcomes and more effective disease prevention.
Thank you to arXiv for use of its open access interoperability. arxiv.org. aibrief.ai is not affiliated with or endorsed by arXiv.