The Role of Machine Learning Algorithms for Diagnosing Diseases

TLDR

Machine learning algorithms are increasingly pivotal in medicine for diagnosing and early predicting diseases from medical databases. This paper provides an overview of machine learning algorithms—Naïve Bayes, logistic regression, SVM, K‑NN, K‑means, decision tree, and random forest—used for disease identification and prediction. The authors reviewed recent studies from the past three years that applied these algorithms to detect diseases, and compared their algorithms, assessment processes, and results. The paper concludes with a discussion of the reviewed works, highlighting their comparative strengths and limitations.

Abstract

Nowadays, machine learning algorithms have become very important in the medical sector, especially for diagnosing disease from the medical database. Many companies using these techniques for the early prediction of diseases and enhance medical diagnostics. The motivation of this paper is to give an overview of the machine learning algorithms that are applied for the identification and prediction of many diseases such as Naïve Bayes, logistic regression, support vector machine, K-nearest neighbor, K-means clustering, decision tree, and random forest. In this work, many previous studies were reviewed that used machine learning algorithms for detecting various diseases in the medical area in the last three years. A comparison is provided concerning these algorithms, assessment processes, and the obtained results. Finally, a discussion of the previous works is presented.