Survey on Classification of Brain Tumor using Wavelet Transform and PNN

Ankit Kashyap, Dr.Amit Agrawal

This paper presents, a new method for Brain Tumor Classification using Probabilistic Neural Network with Discrete Wavelet Transformation is proposed. Human inspection was the conventional method available for computerized tomography, magnetic resonance brain images classification and tumor detection. The classification methods that are operator assisted are impractical incase of large amount of data that are also non reproducible. Operator performance leads to serious inaccuracies in classification by producing noise in Computerized Tomography and Magnetic Resonance images. Neural Network techniques has shown great potential in the field of medical diagnosis. Hence, in this paper the Probabilistic Neural Network with Discrete Wavelet Transform was applied for classification of brain tumors. Classification was performed in two steps, i) Dimensionality reduction and Feature extraction using the Discrete Wavelet Transform and ii) classification using Probabilistic Neural Network (PNN). Evaluation was performed on image data base of Brain Tumor images. The proposed method gives better accuracy when compared to previous methods of classification.