Abstract

Newborn and child mortality prevention is one of the prioritized Sustainable Development Goals (SDG) targets of the World Health Organization (WHO) expected by 2030. This SDG Target 3.2 has prompted for a lot of solutions for the most affected regions like Sub-Saharan Africa, Central and Southern Asia in order to stop preventable deaths of children under the age of 5 years and newborns although available solutions are not yet sufficient to achieve the Goal. The 4th Industrial revolution has further advanced the need for a reliable interdisciplinary approach that leverages technological advancements in achieving the SDG Target 3.2. In this work, we present a trustworthy Artificial Intelligence (AI) solution that blends with the data driven emerging technologies to reduce this global burden by transparently and interpretably predicting preterm births for patients and physicians for predictive and preventive action towards lowering neonatal deaths and increasing child survival. This AI solution can globally improve maternal and child healthcare among nations the run curative healthcare systems. We used Random Forest and KNeighbors and obtained an accuracy of 100% and 78% with respectively with Synthetic Minority Oversampling Technique (SMOTE) and Adaptive Synthetic (ADASYN) class balancing techniques. With interpretability of the random forest algorithm, we can responsibly improve AI technology adoption for maternal and child health and provide useful automated data driven insights to maternal healthcare management stakeholders and policy makers for a sustainable healthcare system in developing countries.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call