IoT-Based Smart Health System for Ambulatory Maternal and Fetal Monitoring

The adoption of IoT for smart health applications is a relevant tool for distributed and intelligent automatic diagnostic systems. This work proposes the development of an integrated solution to monitor maternal and fetal signals for high-risk pregnancies based on IoT sensors, feature extraction based on data analytics, and an intelligent diagnostic aid system based on a 1-D convolutional neural network (CNN) classifier. The fetal heart rate and a group of maternal clinical indicators, such as the uterine tonus activity, blood pressure, heart rate, temperature, and oxygen saturation are monitored. Multiple data sources generate a significant amount of data in different formats and rates. An emergency diagnostic subsystem is proposed based on a fog computing layer and the best accuracy was 92.59% for both maternal and fetal emergency. A smart health analytics system is proposed for multiple feature extraction and the calculation of linear and nonlinear measures. Finally, a classification technique is proposed as a prediction system for maternal, fetal, and simultaneous health status classification, considering six possible outputs. Different classifiers are evaluated and a proposed CNN presented the best results, with the F1-score ranging from 0.74 to 0.91. The results are validated based on the diagnosis provided by two specialists. The results show that the proposed system is a viable solution for maternal and fetal ambulatory monitoring based on IoT.