Smart Monitoring of Maternal Health: An Iot-Ia Hybrid Model For Predicting and Preventing Obstetric Near-Miss

The occurrence of Obstetric Near-Miss (ONM) in pregnant patients is a critical maternal health issue in Colombia, influenced by socio-economic conditions, pre-existing medical conditions, pregnancy-related complications, and barriers to healthcare access. This study aims to design a hybrid IoT-AI model to mitigate the risk of obstetric complications by enabling early risk classification and continuous monitoring of high- and medium-risk patients. A descriptive retrospective study was conducted at Fundación Hospital San Pedro (FHSP) in Pasto, Colombia, analyzing data from 21,770 patients treated between 2013 and 2023, of whom 2,424 were classified as ONM cases. Through data analysis, patterns and relationships between variables associated with ONM were identified, leading to the development of a risk classification model (high, medium, and low) applied to patients attending the hospital. To monitor high- and medium-risk patients, a Kit IoT is being developed, currently in the testing phase, to measure blood pressure, heart rate, oxygen saturation, and temperature, with an interactive screen to assess key clinical symptoms. This paper presents the characterization of pregnant patients in Colombia based on collected data, the description of ONM and its associated factors, the development of the risk classification model, and the design of the proposed IoT solution. The integration of IoT and AI in this system aims to enhance early detection and reduce ONM incidence in Colombia.