Implementing brain tumor detection using various machine learning techniques

Abstract

The brain is a very complex organ of the human body. One of the brain diseases is a tumor. Brain tumors are caused by uncontrolled cell growth. Early recognition, classification and analysis of brain tumors is very important to find out whether there is a tumor in a person's brain so it is important for us to do this in order to treat the tumor thoroughly. Machine learning (ML) techniques that have the highest accuracy in detecting the health sector are extreme gradient boosting (XGBoost), logistic regression, random forest, k-nearest neighbor (KNN), naive Bayes, and support vector machine (SVM). In this research, data collection and exploration were carried out, data training using six methods, and evaluation using a confusion matrix. After conducting the experiment, it was obtained that random forest had the highest accuracy, namely 98.41%. Where XGBoost obtained an accuracy of 98.14%, logistic regression obtained an accuracy of 97.34%, KNN and naive Bayes of 97.34%, and SVM of 97.88%.