Identification of Glioma using Discrete Wavelet Transform (DWT) and Artificial Neural Network (ANN)

Endah Nur Salamah, Dian C Rini Novitasari, Ahmad Hanif Asyhar, Muh Ma'arif

2018

Abstract

Glioma is one of the deadly diseases and suffered by many people in the world. Glioma means brain tumor. In 2016, the World Health Organization (WHO) recorded as many as 6.2 million people of the world suffering gliomas. Based on this fact, it is necessary to examine glioma using a tool that is one of magnetic resonance imaging (MRI), then the results of brain MRI image be analyzed or diagnosed by an expert doctor but sometimes the results of its analysis is still subjective and takes a long time. The image used in this study are the normal brain MRI image and glioma brain MRI image. First steps are image improvement (Adaptive Histogram Equalization), second step is image segmentation using otsu threshold, third step is image extraction using discrete wavelet transform (DWT) with the features took are energy, standard deviation and mean, then classification using ANN (backpropagation) which was identified into two classes, normal and glioma. The result of classification using backpropagation with its structure that has been determined, that a is 25 nodes hidden layer, 4 levels decomposition, obtained MSE error = 0,0000999, sensitivity value 100%, specificity equal to 85,71%, and accuracy equal to 91,67%.

Paper Citation

in Harvard Style

Salamah E., Novitasari D., Asyhar A. and Ma'arif M. (2018). Identification of Glioma using Discrete Wavelet Transform (DWT) and Artificial Neural Network (ANN). In Proceedings of the Built Environment, Science and Technology International Conference - Volume 1: BEST ICON, ISBN 978-989-758-414-5, pages 289-294. DOI: 10.5220/0008906700002481

in Bibtex Style

@conference{best icon18,
author={Endah Nur Salamah and Dian C Rini Novitasari and Ahmad Hanif Asyhar and Muh Ma'arif},
title={Identification of Glioma using Discrete Wavelet Transform (DWT) and Artificial Neural Network (ANN)},
booktitle={Proceedings of the Built Environment, Science and Technology International Conference - Volume 1: BEST ICON,},
year={2018},
pages={289-294},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0008906700002481},
isbn={978-989-758-414-5},
}

in EndNote Style

TY - CONF

JO - Proceedings of the Built Environment, Science and Technology International Conference - Volume 1: BEST ICON,
TI - Identification of Glioma using Discrete Wavelet Transform (DWT) and Artificial Neural Network (ANN)
SN - 978-989-758-414-5
AU - Salamah E.
AU - Novitasari D.
AU - Asyhar A.
AU - Ma'arif M.
PY - 2018
SP - 289
EP - 294
DO - 10.5220/0008906700002481