Pathological Brain Detection Using Weiner Filtering, 2D-Discrete Wavelet Transform, Probabilistic PCA, and Random Subspace Ensemble Classifier

Pseudocode 1

Pseudocode of the proposed system.

Input: T2-weighted MR brain images.

Parameter: , total number of images

Step  1 (weiner filter)

for

Read the images and apply wiener filter

end

Step  2 (2D-DWT)

For

Read in the image file

Apply the DWT using for the 3rd level using “Haar” wavelet to extract the wavelet coefficients.

A matrix [] is employed to store all the coefficients.

End

Step  3. Reduce the features from the coefficients using PPCA

for

Apply PPCA transformation on the obtained wavelet coefficients.

Put the new dataset in a matrix .

End

Step  4 (RSE classification using 5 5 cross-validation)

Divide the input data and target data into 5 different groups randomly

For

Use the th group for test, and other 4 groups to train the RSE algorithm.

Classify test image

End

Calculate average specificity, sensitivity, and accuracy.