Iod-Nets – An IoT based intelligent health care monitoring system for ambulatory pregnant mothers and fetuses

Abstract

Internet of things (IoT) and Artificial Intelligence (AI) have become the most predominant tools in healthcare applications for pervasive and smart systems for automated diagnosis. Based on the integration of IoT sensors and deep learning algorithms, this study proposes the development of intelligent monitoring systems for maternal and foetal signals in high-risk pregnancies. IoT sensors collect maternal clinical data, such as temperature, blood pressure, oxygen saturation level, heart rate, and fetal heart rate and store them on the cloud for monitoring and prediction purposes. Furthermore, a novel Optimized single-dimensional Convolutional Neural Network (1D-OCNN) is proposed for better classification and prediction of the different emergencies of both mother and fetus. The IoT systems have been designed based on the multiple sensors interfaced with the MICOT boards (Node MCU+MCP3008) and cloud mechanism. Nearly 9000 data were collected and used for the evaluation. The extensive experimentation is carried out using cloud-centric learning algorithms such as K-Nearest neighborhood (KNN), Random forest(RF), Support vector machines(SVM), Convolutional neural networks(CNN) and Extreme learning machines (ELM) and various parameters such as accuracy, precision, recall, sensitivity, and F1-score are calculated. The evaluation shows that the proposed classifier has outperformed the other learning algorithms regarding accuracy, precision, recall and F1-score. Based on the above results, the suggested system is a practical and efficient alternative for maternal and fetal monitoring using IoT and artificial intelligence.