A Comprehensive Study of Disease Prediction using Machine Learning
Abstract - Machine Learning (ML) extensively used and popular in the Healthcare sector for the last few decades. Analysis of complex structured EHRs manually is time-consuming. To process EHRs automatically there is a need for Machine learning. The objective of this work is to examine the current state-of-the-art followed by ML Algorithms, Tools, and Performance Metrics. In this work, we have collected research work from various sources related to several chronic disease predictions by ML. This work concluded that (1) Heart Disease and Diabetes are the leading cause of death all around the world; (2) SVM is the most popular algorithm for Disease Prediction; (3) Most of the Experiments carried out by WEKA Software; (4) For evaluation of Algorithm(s) performance metrics are Precision, Recall, F1-Measure, Accuracy, AUC and the most used parameter is Accuracy. These insights will aid researchers and healthcare institutions working in this field in broadening their knowledge base.
Keywords - Disease, Machine Learning, Logistic Regression, Support Vector Machine, Neural Network