MRI based brain tumor detection using wavelet packet feature and artificial neural networks

  • Authors:

  • R. Mishra

  • Affiliations:

  • Indraprastha University, Delhi, India

  • Venue:
  • Proceedings of the International Conference and Workshop on Emerging Trends in Technology
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

Magnetic Resonance Imaging (MRI) is a non-invasive technique for assessing abnormalities of tissue composition. The objective of the paper is to develop an expert system which can diagnose Brain Tumor with the highest accuracy using artificial neural network. After extracting features from MRI data using wavelet packets, we use artificial neural networks to determine the abnormal and normal spectra. The advantage of wavelet packets is that it gives richest analysis when compared with the wavelet transforms there by adding advantages to the performance of the system. Two cancer detection Techniques have been discussed in this paper based on which the conclusion is drawn. Error Back Propagation Training Learning rule is used to train the neural network system.