Smart Neonatal Guardian: AI-Driven Monitoring for Preterm Infants in NICU

Abstract

Minimally designed, Neonatal Intensive Care Units (NICUs) are specialized medicinal settings that offer the highest level of care to preterm and ill infants requiring 24/7 assessment for pharmacological intervention. Monitoring of vital parameters (heart rate, oxygen saturation [SpO2], and body temperature) is crucial for providing safety and stability to these infants. Conventional NICU monitoring systems are costly and static components isolated within the hospital infrastructure, which cannot access real-time data outside clinical contexts.

Thesis synopsis: This study describes the design of a neonatal monitoring system based on machine learning and Internet of Things (IoT) that continuously measures vital signs in neonates. The proposed method consists of utilizing sensors to collect data including heart rate, blood oxygen saturation levels and body temperature that are attached to an ESP32 microcontroller for acquisition and processing. It is a device whose sensing data are collected and wirelessly transferred using Blynk platform to a mobile application.

These abnormal physiological value triggers alert the system through a built-in buzzer and mobile notifications, so that individuals can attend to patients who need medical care. The proposed system is cost-effective and portable solution which not only enhance traditional monitoring techniques but also provide a way for continuous monitoring of patient data with access from remote regions. This helps facilitate timely intervention and communication between healthcare providers and caregivers, ultimately improving neonatal health.