Maternal risk prediction is a critical component of prenatal care aimed at mitigating adverse outcomes for pregnant individuals and their babies. In this research, we have employed machine learning algorithms to forecast maternal risks using a comprehensive dataset that has 3097 data points and 9 features. Eight algorithms showed notable different accuracies -AdaBoost 80.24%, XGBoost 93.97%, GBoost 88.84%, SVC 73.52%, KNN 84.85%, Naive Bayes 70.94%, Decision Tree 91.93%, Random Forest 94.15%. Here XGBoost and Random Forest achieved the highest accuracies of 93.97% and 94.15% respectively, with high precision and recall. These findings develop machine learning in maternal risk prediction with highlighting the potential for improving prenatal care through personalized risk assessment and targeted interventions.