The Performance of Brain Tumor Diagnosis Based on Machine Learning Techniques Evaluation - A Systematic Review

This research aims to investigate the performance of brain tumor diagnosis and treatment using machine learning algorithms. This study provides systematic review of papers on the improvement of human life. The papers reviewed are taken from relevant articles published between October 2012 and December 2019. The investigation is done against the algorithm type, dataset, the proposed model, and the performance in each of the papers. The accuracy result among the papers papers studied is ranged between 79% - 97.7%. The algorithms they used are CNN, KNN, C-means, RF, respectively, ordered from the highest frequency of use to the lowest. In the papers studied, it was shown that various methods had been used with good results. However, the confidence in the research results in term of accuracy for the detection of brain tumors still needs to be increased. Furthermore, building a software applications can be very useful to solve real cases.