Using AI-Based Approaches in Health Care for Predicting Health Issues in Pregnant Women

In developing countries, healthcare monitoring is a crucial component for determining the well-being of the patient, and at the same time, it is highly effective for managing the hospital resources which are limited in developing countries like India. With the increase in amount of digital data available for healthcare sector and the development of tools for data analysis, a new era of medical treatment could be arrived at with stringent methods of checking the efficacy of medicines and devising statistically valid tests that would authoritatively show the scope of medical treatment. This paper proposes a conceptual healthcare data model based on data mining techniques. The predictions have been made based on NHRM and RCH dataset, and three data mining tools, i.e., IBM SPSS Modeler, RapidMiner, and Weka, are used for determining the performance of the proposed model. Different classifiers, namely CHAID, random forest, K-nearest neighbor, logistic regression, decision tree, Naive Bayes, and C5.0, are used for predicting the stay of pregnant ladies in the hospital. A detailed analysis of these classifiers on the data mining toolset is also presented in this paper.