Maatran: Revolutionizing Maternal Care Through Remote Monitoring and Risk Prediction

Maternal mortality due to pregnancy complications remains a significant global challenge, necessitating the development of innovative approaches to safeguard pregnant women from potential threats. This paper introduces Maatran, a technological solution that leverages IoT, Cloud technologies, and Machine Learning solutions to address maternal health issues. Maatran utilizes Arduino-based wearable sensor devices to remotely collect health data, which is then transmitted to an Android app interface for data processing and storage in the cloud. A Random Forest model is employed to accurately predict maternal health risks. The system’s features provide a holistic solution for remote health monitoring of pregnant women and facilitate direct communication with healthcare professionals. Through rigorous machine learning model training, Maatran achieved accuracy, ROC-AUC, and log loss of 0.93, 0.91, and 0.93 respectively, surpassing existing state-of-the-art models. The implementation of Maatran demonstrates its potential to significantly improve maternal health outcomes by enabling early risk detection and timely interventions. The supplementary document can be found at https://github.com/kulsumkamal/Maatran-Supplementary .