Data Mining Techniques in Diagnosis of Chronic Diseases

Chronic diseases and cancer are raising health concerns globally due to lower chances of survival when encountered with any of these diseases. The need to implement automated data mining techniques to enable cost-effective and early diagnosis of various diseases is fast becoming a trend in healthcare industry. The optimal techniques for prediction and diagnosis vary between different chronic diseases and the disease related-parameters under study. This review article provides a holistic view of the types of machine learning techniques that can be used in diagnosis and prediction of several chronic diseases such as diabetes, cardiovascular and brain diseases, chronic kidney disease and a few types of cancers, namely breast, lung and brain cancers. Overall, the computer-aided, automatic data mining techniques that are commonly employed in diagnosis and prognosis of chronic diseases include decision tree algorithms, Naïve Bayes, association rule, multilayer perceptron (MLP), Random Forest and support vector machines (SVM), among others. As the accuracy and overall performance of the classifiers differ for every disease, this article provides a mean to understand the ideal machine learning techniques for prediction of several well-known chronic diseases.