There are remarkable improvements in the healthcare sector particularly in patient care, maintaining and protecting the data, and saving administrative and operating costs, etc. Among the various functions in the healthcare sector, disease diagnosis is considered as the foremost function because it saves a life at the correct time. Early detection of diseases helps in disease prevention, letting the patients get vigorous and effective treatment and saving their lives. Several techniques were suggested by the researchers for disease prediction. Many literatures have been witnessed on disease prediction. This article reviews several articles systematically and compares various machine learning (ML) algorithms for disease prediction, including the Random Forest (RF), Naive Bayes (NB), Decision Tree (DT), Support Vector Machine (SVM), and Logistic Regression (LR) algorithms. A thorough analysis is presented based on the number of publications year-wise, disease-wise, and also based on the performance metrics. This review thoroughly analyzes and compares various ML techniques applied in disease prediction, focusing on classification algorithms commonly employed in healthcare applications. From the systematic review, a multi objective optimization method named Grey Relational Analysis (GRA) is used to rank the ML algorithms using their performance metrics. The results of this paper help the researchers to have an insight into the disease prediction domain. Also, the performance of various ML algorithms aids the researchers to choose a better methodology to predict a disease.