An essential part of public health, maternal healthcare looks after the expectant mother’s health prenatal, perinatal, postnatal. It guarantees the best possible health for the growing fetus as well as the mother through its predictions. Maternity and the postpartum period present a multitude of risks and repercussions for the mother’s health, and hence prompt identification of these risks can be extremely important to a woman’s safety. A strategy to anticipate hazards related to maternal health is proposed in this study. Principal component analysis (PCA) is the initial step in the process, which involves extracting important features from the dataset. To attain good performance, this study then uses a stacked ensemble voting classifier that includes one deep learning model and one machine learning model. By contrasting it with current state-of-the-art methods, the suggested model’s efficacy is further verified. Comparison analysis was made with many Machine Learning (ML) models such as Random forest classifier, Logistic Regression, Decision tree classifier, Support vector classifier, and XGB classifier. Based on a comprehensive investigation, the suggested deep learning model (one Machine Learning and one Deep Learning) yields superior prediction rates for high and low risk, with 84% and 86% of accuracy, respectively.