DEVELOPMENT OF SMART TEXTILE FOR INFANT HEALTH MONITORING BY INTEGRATING IoT AND MACHINE LEARNING
Mounika C, Ciratheep P, Dharshini C, Muhammed Zaid, Shanmathi P
Keywords: Biomedical textiles, conductive embroidery, infant monitoring, temperature sensors, wearable electronics.
Abstract
Smart fabrics are proving to be an exciting device in pediatric care by facilitating continuous, non invasive physiological monitoring of infants. This paper is concerned with the fabrication of a cotton based smart fabric wristband interfaced with digital temperature and PPG-based heart rate/SpO₂ sensors to establish practical, real-time monitoring. The selected cotton substrate provides softness, breathability, and biocompatibility for sensitive infant skin, while conductive embroidery with silver coated yarn provides a robust electronic pathway without affecting fabric comfort or wear. A low-power microcontroller with Wi-Fi communication was utilized for data sampling and wireless uploading to a dedicated mobile app, which allows real-time visualization, threshold-based alarms, historical data analysis, and secure cloud storage. The integration of textile and sensor was tested using fabric elasticity, wash durability, skin compatibility, and pressure distribution tests to ensure its feasibility for extended use in infants. Sensor performance was compared against clinical reference devices to ensure measurement accuracy and consistency. In addition to real-time observation, the system also utilizes machine learning algorithms, such as logistic regression, random forest, long short-term memory (LSTM) neural networks, and anomaly detection algorithms, to facilitate predictive healthcare through the detection of early physiological shifts and emergent health threats. The research identifies the potential use of integrating smart biomedical fabrics with IoT-capable data transfer and predictive analytics for reversing infant health monitoring from passive measurement to active care.