Secure IoT based smart system for monitoring health care for ambulatory and fetal

Abstract

The Internet of Things (IoT) along with Artificial Intelligence (AI) has now developed the most prevalent instruments in applications in the health care industry for widespread and intelligent automatic diagnostic systems. The maternal clinical information for ubiquitous and intelligent autonomous diagnostic systems, IoT with AI has now taken the lead as the predominant instrument in the healthcare industry. With respect to the incorporation of IoT sensors along with deep learning techniques, this article suggests the establishment of smart networks to monitor maternal and fetal signs in dangerous pregnancies. IoT sensors are utilized to gather clinical information about the mother, including her temperature, blood pressure levels, and saturation in oxygen level, heart rate, and heartbeat of the unborn child. This information is then stored in the cloud for tracking and forecasting. Additionally, a brand-new optimal Gated Recurrent Unit (GRU) is suggested for improved categorization and a forecast of the several emergencies affecting both pregnant women and unborn children. Additionally, to assist the data’s security and the design of IoT systems is based on a number of sensors that are interfaced with MICOT boards (Node MCU+MCP3008) and a cloud system. For the evaluation, about 500 pieces of data were gathered and utilized. With cloud-centric learning techniques like K-Nearest Neighbor (KNN), Random Forest (RF), Support Vector Machines (SVM), Convolution Neural Networks (CNN), and Extreme Learning Machines (ELM) thorough experimentation is conducted, and various parameters including accuracy, precision, recall, and sensitivity, along with F1-score are estimated. The evaluation found that the recommended classifier outperformed the competing learning strategies. The recommended framework is a practical and useful method for maternal and foetal surveillance that is powered by IoT and AI.