A Systematic Literature Review on Machine Learning in Healthcare Prediction

Abstract

Rapid technological advancement will continue to create new values and transform experiences in many sectors, including healthcare. Several key trends are shaping today’s healthcare system, including the use of machine learning (ML). This systematic literature review (SLR) explores the application of ML in healthcare, particularly in predictive analytics. The SLR also includes a few papers on machine learning operations (MLOps) in healthcare, reflecting limited studies on the topic. This suggests significant potential for further exploration in MLOps. The review compares findings from various studies, many of which agree that ML enhances the scalability and reliability of predictive models. This study aims to assess the most effective ML algorithms and methodologies used in healthcare prediction. It also attempts to identify features influencing the outcomes of ML applications in healthcare predictions. Findings suggest that ML can improve prediction accuracy using the appropriate dataset, optimal feature selection model, and a tailored ML algorithm for specific tasks. The literature highlights challenges, including the need for specialised skills and the complexity of integrating MLOps into existing healthcare systems.