Subject Area: COMPUTER SCIENCE
The paper presents a review of machine learning approaches for the early detection of Stress and Hypertension (SH) in pregnant women. It discusses the management of SH in pregnant women and its causes, such as psychosocial stressors, anxiety and depression, pregnancy-related concerns, physiological changes, gestational diabetes, poor lifestyle, complications, and medical concerns. Existing methods for managing SH among pregnant women, such as stress questionnaires, psychological assessments, urine protein testing, symptom assessment, blood pressure monitoring, fetal monitoring, biochemical markers, and ambulatory blood pressure monitoring, are also presented. Furthermore, the paper discusses the challenges encountered in adopting these conventional methods, including the subjectivity of self-reported data, limited access to healthcare services, diagnostic challenges, cost and resource constraints, interpretation and variability, patient compliance and engagement, and the complexity of multifactorial conditions. The study suggests that addressing these challenges requires efforts to improve healthcare infrastructure, increase access to prenatal care services, enhance diagnostic capabilities, and promote patient education and engagement. By addressing these barriers, healthcare systems can better detect and manage SH in pregnant women, ultimately improving maternal and fetal outcomes. Additionally, the paper explores the implementation of various intelligent methods based on machine learning and deep learning for the early detection of stress and hypertension. It reviews various works on the implementation of intelligent techniques, which have shown more satisfying and adaptive results. However, despite the extensive discussion on the impact and management of SH among pregnant women, the paper notes the absence of a solution for a cost-effective software model for the management of SH among pregnant women.