Ai-Based N8n Workflows For Professional Support In Ambulatory Glucose Monitoring: A Proof of Concept

This paper presents a proof of concept based on n8n workflows supported by artificial intelligence for professional assistance in ambulatory glucose monitoring. The proposal addresses a common gap in outpatient diabetes follow-up: the difficulty of transforming patient-generated data into timely, traceable, and clinically useful actions. The proof of concept integrates workflow orchestration, structured event logging, adherence-oriented processing, and human-in-the-loop professional review supported by large language models as draft generators rather than autonomous decision-makers. The implemented workflows were evaluated with simulated patient data to demonstrate operational feasibility, traceability, and measurable process outputs. The results show that the proposed approach can support the generation of review tasks, organize relevant monitoring events, and assist professionals in follow-up activities while preserving human oversight. This work does not claim clinical efficacy; instead, it demonstrates the technical and operational viability of an architecture for safer and more measurable ambulatory glucose follow-up.