Published Paper Details:

The Mud Ring Algorithm (MRA) is used in the study to improve the performance of machine learning classifiers in maternal health risk prediction. The technique is applied in two steps: first, MRA increases the predicted accuracy of the Support Vector Machine (SVM) by optimizing its parameters. Multiple datasets are used to assess the model, and it is contrasted with current optimization methods. A crossover oversampling strategy is used in the second stage to rectify the class imbalance in the maternal health risk dataset. K-Nearest Neighbor and Random Forest are two more classifiers that are incorporated to increase prediction accuracy. The suggested method considerably improves classification performance, according to experimental data, with MRAbased optimisation raising accuracy by 11.8% for SVM, 9.11% for Random Forest, and 17.08% for KNN. In order to follow the health of mothers and children, a digital health monitoring system is also introduced. In addition to gathering health data, the technology forecasts hazards including preeclampsia and gestational diabetes and offers real-time insights. Following delivery, it keeps track of immunisations and development indicators, produces digital birth certificates, and documents important infant information. Authorised users can effectively monitor and manage healthcare because data security and accessibility are guaranteed. This strategy streamlines administrative procedures while enhancing maternal and child health outcomes by fusing predictive analytics with automated healthcare management.