Rule-Based Severity-Tiered Alert Routing in a Simulation-Based Cloud–IoT Neonatal Monitoring Pipeline: An Engineering Proof of Concept

Neonatal intensive care requires continuous vital-sign monitoring and structured notification workflows. This study presents a layered, simulation-based Cloud–IoT engineering proof of concept linking controlled physiological inputs to rule-based alert routing. The framework generates five physiological inputs, transmits JSON-formatted messages through MQTT and Azure IoT Hub, stores the records, and visualises them through Power BI dashboards. A transparent Rule-Based Engineering Severity Index (0–5) counts physiological categories meeting configurable operational rules and determines notification routing through Telegram, email, and Twilio voice calls. The index supports workflow orchestration and is not a clinically validated neonatal severity score. Evaluation included four controlled workflow demonstrations and repeated notification-service transactions through Telegram (n = 30), email (n = 30), and Twilio call creation (n = 10). A separate acknowledgment experiment showed detection in all 10 intentional ACK trials, while all five intentional NO-ACK trials triggered additional email submission after timeout expiration. Additional software-level tests examined transient-abnormality filtering, malformed or missing data, and local buffering with retransmission after restored connectivity. These evaluations support the feasibility of a traceable monitoring and notification workflow under the tested simulated conditions. Clinical effectiveness, infrastructure resilience, and scalability remain outside the demonstrated scope.