A Comparative Analysis on Random Forest Algorithm Over K-Means for Identifying the Brain Tumor Anomalies Using Novel CT Scan with MRI Scan

Abstract

The aim of the work is to identify the brain tumor anomalies by using CT scan with MRI scan. Two machine learning algorithms Random forest algorithm and K-means are used to classify CT scan with MRI images. To achieve maximum accuracy, the sample size n=5 in Random forest and n=5 in Kmeans was iterated 10 times for efficient and accurate analysis on MRI images with G power in 80% and threshold 0.05%, CI 95% mean and standard deviation. The experimental results show that the Random forest algorithm with mean accuracy of 84% is compared with the K-means classifier algorithm of mean accuracy 72%. There is a statistically significant difference between the study groups with (P<0.05). Based on the results achieved, the Random Forest classification algorithm better identifies brain tumor anomalies than the K-means classifier algorithm.