A Comprehensive Review of Machine Learning Methods for Healthcare-Related Illness Prognosis

Abstract

There has been an explosion of healthcare data in the 21st century, along with an increase in chronic illnesses and other health-related issues. Conventional data management solutions struggle to keep up with the sheer volume of healthcare data. The need for intelligent decision support systems that enable doctors to make better decisions via the use of various algorithms brings machine learning into play. This paper summarizes the many kinds of healthcare data and analyses the results of many studies according to how well machine learning approaches forecast illnesses and identify high-risk diseases early on. The most often mentioned Machine Learning techniques were Support Vector Machine (SVM), Random Forest, Decision Trees (DT), K Nearest Neighbor (KNN), Deep Learning, Naive Bayes (NB), Artificial Neural Network (ANN), and Logistic Regression. Primarily considered are algorithms, datasets, accuracy, and knowledge gaps in the field.
Index Terms –Wellness program, Machine Learning, Disease Prediction, Heterogeneous data.