Jordan University of Science and Technology

A Classifier to Detect Tumor Disease in MRI Brain Images

Authors:  Amer Al-Badarneh, Hassan Njadat, and Ali Razqi

The traditional method for detecting the tumor diseases in the human MRI brain images is done manually by physicians. Automatic classification of tumors of MRI images requires high accuracy, since the non-accurate diagnosis and postponing delivery of the precise diagnosis would lead to increase the prevalence of more serious diseases. To avoid that, an automatic classification system is proposed for tumor classification of MRI images. This work shows the effect of neural network (NN) and K-Nearest Neighbor (K-NN) algorithms on tumor classification. We used a benchmark dataset MRI brain images. The experimental results show that our approach achieves 100% classification accuracy using KNN and 98.92% using NN.