MRI Probabilistic Neural Network Screening System: a benign and malignant recognition case study

Sami Hasan; Mays Yousif; Talib M. J. Al-Talib

doi:10.24996/ijs.2021.SI.1.22

Authors

Sami Hasan College of Information Engineering , Al-Nahrain University , Baghdad, Iraq
Mays Yousif College of Information Engineering , Al-Nahrain University , Baghdad, Iraq
Talib M. J. Al-Talib College of Information Engineering , Al-Nahrain University , Baghdad, Iraq

DOI:

https://doi.org/10.24996/ijs.2021.SI.1.22

Keywords:

Benign Tumor, Brain tumor, Curvelet Transform, Malignant Tumor, Neural Network

Abstract

This work is aimed to design a system which is able to diagnose two types of tumors in a human brain (benign and malignant), using curvelet transform and probabilistic neural network. Our proposed method follows an approach in which the stages are preprocessing using Gaussian filter, segmentation using fuzzy c-means and feature extraction using curvelet transform. These features are trained and tested the probabilistic neural network. Curvelet transform is to extract the feature of MRI images. The proposed screening technique has successfully detected the brain cancer from MRI images of an almost 100% recognition rate accuracy.