An Efficient Interaction Human-AI Synergy System Bridging Visual Awareness and Large Language Model for Intensive Care Units
🤖 AI Summary
High data error rates and excessive clinical workload in ICUs stem from manual transcription and fragmented, isolated information systems. To address this, we propose a cloud-edge-end collaborative human–AI system featuring a novel ICU interaction paradigm that deeply integrates vision-based perception (YOLO + OCR) with a fine-tuned large language model (LLM). This enables contactless, real-time extraction of bedside monitor data (end-to-end latency <200 ms) and semantic voice-based retrieval of fragmented clinical information (response time <1.5 s). Leveraging edge-lightweight deployment and a unified voice interface, the system significantly reduces documentation errors and cognitive load on clinicians. Clinical validation was conducted in the ICU of a tertiary hospital, demonstrating robust performance, scalability, and minimal workflow disruption. Our approach establishes a practical, low-intrusion technical pathway toward intelligent critical care monitoring.
Application Category
📝 Abstract
Intensive Care Units (ICUs) are critical environments characterized by high-stakes monitoring and complex data management. However, current practices often rely on manual data transcription and fragmented information systems, introducing potential risks to patient safety and operational efficiency. To address these issues, we propose a human-AI synergy system based on a cloud-edge-end architecture, which integrates visual-aware data extraction and semantic interaction mechanisms. Specifically, a visual-aware edge module non-invasively captures real-time physiological data from bedside monitors, reducing manual entry errors. To improve accessibility to fragmented data sources, a semantic interaction module, powered by a Large Language Model (LLM), enables physicians to perform efficient and intuitive voice-based queries over structured patient data. The hierarchical cloud-edge-end deployment ensures low-latency communication and scalable system performance. Our system reduces the cognitive burden on ICU nurses and physicians and demonstrates promising potential for broader applications in intelligent healthcare systems.
Problem
Research questions and friction points this paper is trying to address.
Automates ICU data capture from monitors to reduce manual errors
Integrates fragmented patient data via voice queries using LLM
Deploys cloud-edge system for low-latency healthcare communication
Innovation
Methods, ideas, or system contributions that make the work stand out.
Cloud-edge-end architecture for low-latency ICU data processing
Visual-aware edge module captures real-time physiological data non-invasively
LLM-powered semantic interaction enables voice-based queries over patient data
🔎 Similar Papers
No similar papers found.
