The rapid expansion of digital healthcare has necessitated systems that go beyond static appointment booking to offer dynamic, real-time patient monitoring and intelligent clinical triage. This research proposes 'Edge-Heal,' a comprehensive, full-stack healthcare ecosystem that integrates Internet of Things (IoT) sensor data, Artificial Intelligence (AI), and geospatial routing algorithms to optimize patient care and clinical workflows. The system captures real-time vital signs—specifically Heart Rate, Peripheral Oxygen Saturation (SpO2), and Body Temperature—using an array of simulated ESP32 microcontrollers mimicking MAX30102 and LM35 sensors. This high-frequency data is streamed via a high-performance Spring Boot RESTful API backend and visualized dynamically on a React.js monitoring dashboard using Recharts. Concurrently, an integrated Large Language Model (OpenAI GPT-3.5/4) performs automated symptom triage by evaluating patient-reported conditions in natural language. The AI engine classifies symptoms into specific medical domains and drafts preliminary prescriptions, significantly reducing the administrative and diagnostic workload on physicians. Furthermore, to address emergency routing and optimize facility utilization, the system employs the Haversine formula to map and sort 100 hospital nodes across Tamil Nadu, directing patients to the nearest available specialist based on their real-time GPS coordinates. Experimental results demonstrate a 40% reduction in initial triage time, a 99.8% uptime in continuous IoT data ingestion, and highly accurate proximity-based hospital mapping with sub-50ms latency. Edge-Heal provides a highly scalable, end-to-end framework that bridges the critical gap between remote hardware monitoring and AI-assisted clinical decision-making, paving the way for next-generation smart healthcare infrastructure.

Keywords— IoT Healthcare; Artificial Intelligence; Geospatial Triage; Remote Patient Monitoring; Spring Boot; React; Haversine Formula.