RECOVER: Designing a Large Language Model-based Remote Patient Monitoring System for Postoperative Gastrointestinal Cancer Care

2025-02-09 · Source: cs.AI updates on arXiv.org · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Human-Computer Interaction, Medical Devices & Health Technology · Depth: Expert, quick

Summary

RECOVER is a large language model (LLM)-powered remote patient monitoring (RPM) system designed for postoperative gastrointestinal (GI) cancer care, a critical area given that GI cancers account for over 35% of cancer-related deaths worldwide. The system integrates an LLM-powered conversational agent for patients and an interactive dashboard for clinical staff to facilitate efficient postoperative monitoring. Its design process involved seven participatory design sessions with five clinical staff and interviews with five cancer patients, leading to six major design strategies for incorporating clinical guidelines and information needs. A pilot assessment of RECOVER with four clinical staff and five patients provided insights into crucial design elements, responsible AI considerations, and future opportunities for LLM-powered RPM systems.

Key takeaway

For AI Engineers and Research Scientists developing healthcare solutions, RECOVER demonstrates a structured approach to integrating LLMs into remote patient monitoring. You should prioritize participatory design with both clinical staff and patients to ensure your system addresses real-world needs and adheres to clinical guidelines, thereby enhancing patient safety and care efficiency. Consider how your LLM-powered system can offer both patient-facing conversational support and clinician-facing data visualization.

Key insights

LLM-powered RPM systems can enhance postoperative GI cancer care through conversational agents and staff dashboards.

Principles

• Engage stakeholders in design.
• Integrate clinical guidelines.
• Prioritize responsible AI.

Method

The design process for RECOVER involved participatory design sessions with clinical staff and patient interviews to derive six key strategies, followed by system implementation and pilot assessment to refine design implications.

In practice

• Use conversational agents for patient support.
• Develop interactive dashboards for clinicians.
• Conduct pilot assessments with real users.

Topics

• RECOVER System
• Large Language Models
• Remote Patient Monitoring
• Gastrointestinal Cancer Care
• Participatory Design

Best for: AI Scientist, AI Engineer, Research Scientist

Related on AIssential

• From 'What' to 'How' and 'Why': Sharing LLM-Generated Retrospective Summaries of Older Adults' Passive Tracking Data with Remote Family Members — LLM-generated summaries for remote family members require context-aware narratives beyond raw data to be helpful and trusted.
• PSEBench: A Controllable and Verifiable Benchmark for Evaluating LLMs in Patient Safety Event Triage — LLM evaluation for high-stakes patient safety triage requires policy-grounded, interactive benchmarks beyond static classification.
• MedRLM: Recursive Multimodal Health Intelligence for Long-Context Clinical Reasoning, Sensor-Guided Screening, Evidence-Grounded Decision Support, and Community-to-Tertiary Referral Optimization — MedRLM recursively processes heterogeneous patient data as an external clinical environment for auditable, workflow-aware decision support.
• When LLMs Analyze Scars: From Images to Clinically-Meaningful Features — LLMs can act as knowledge-driven feature engineers, generating code for clinically interpretable image features to overcome data scarcity.
• LLM-Powered Personalized Glycemic Assessment in Type 2 Diabetes with Wearable Sensor Data — LLMs effectively integrate diverse data for personalized T2D glycemic assessment, surpassing traditional ML methods.
• Clinically Interpretable Sepsis Early Warning via LLM-Guided Simulation of Temporal Physiological Dynamics — LLM-guided temporal simulation improves sepsis early warning by providing interpretable physiological trajectory predictions.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by cs.AI updates on arXiv.org.