Application of Machine Learning and Predictive Models in Healthcare – A Review

Abstract

The use of predictive analytics or models in healthcare has the potential to revolutionize patient care by identifying high-risk patients and intervening with targeted preventative measures to improve health outcomes. This makes the application of analytics in healthcare a concept of utmost interest, which has been explored in various fashions by several scholars. From predicting patients’ ailments to prescribing appropriate drugs, predictive models have seen massive interest. This work studied published works on predictive models in healthcare and observed that the implementation of predictive models in healthcare is experiencing a notable upswing, with a particular focus on research in the United States, where a majority of the top publications originated. Surprisingly, all of the leading nations in this sector have affiliations spanning many continents, with the exception of Africa and South America, together producing a substantially larger volume of research than other countries. The United States also shone out, accounting for 60% of the top five researchers. Notably, although it was published in 2017 (relatively later), Jiang et al. had the most citations (1,346). These studies' core themes were clinical standards, machine learning terminology, and model accuracy. The Journal of Biomedical Informatics topped among journals, with 54 articles, while Luo Gang emerged as the top-performing author, with 12 publications.