A Systematic Study About Chronic Disease Prediction Strategies using Artificial Intelligence based Algorithms

The challenge of chronic diseases will take its deserved place among the top global health issues, mainly because they account for 74% of deaths every year, as reported by the WHO. This systematic review will discuss possible applications of machine learning and AI techniques to predict, diagnose, and treat CDs such as diabetes, cancer, cardiovascular diseases, and liver disease. Integration of ML algorithms: RF, SVM, DT, ANN, and boosting techniques has transformed healthcare by improving the quality of predictive information for potential cases, reducing healthcare costs, and contributing to innovative clinical decision support systems. The research focuses on several approaches of ML, namely supervised and unsupervised learning, for predicting patient-specific treatment regimens and hence better outcomes in chronic disease management. Another discussion from the study is on how to apply ML on large, structured and unstructured datasets collected from the healthcare system about the predictive powers of AI-based models in order to be able to address complications arising from CD. Research aims at providing a comprehensive overview of the understanding of predictive models (PMs) under one heading, making an amalgamation of the desired workings of PMs for chronic disease prediction, useful for knowledge collection for future reference. The paper also provides a descriptive analysis of data sources, methodologies, and algorithms adopted, accompanied by a tabular representation of previous studies to avoid redundancy and enhance accuracy.