Research Article

Classification of Region of Interest in Mammograms Using Dual Contourlet Transform and Improved KNN

Table 3

Classification performance (%) using KNN and improved KNN tested between normal and abnormal.

FeatureDual-CTContourletWavelet
NormalAbnormalNormalAbnormalNormalAbnormal

KNN (mean)84.4779.6687.3853.3988.3561.02
Improved KNN (mean)96.1288.1498.5480.5195.1581.36

KNN (SD)85.4473.7383.9861.8687.8662.71
Improved KNN (SD)96.6088.1497.5785.5994.1787.29

KNN (smoothness)83.9877.9790.2961.0287.8667.80
Improved KNN (smoothness)94.6692.3796.6081.3690.7887.29

KNN (skewness)83.9877.1286.4155.0883.9059.32
Improved KNN (skewness)95.1588.9893.6983.0591.7584.95

KNN (uniformity)85.9262.7188.8361.3684.4766.95
Improved KNN (uniformity)98.5484.7596.6087.2996.6084.75

KNN (entropy)84.9570.3487.3858.4482.5263.56
Improved KNN (entropy)98.0687.2999.0383.9097.5786.44

KNN (contrast)83.9867.8089.8153.3980.5866.10
Improved KNN (contrast)96.6080.5198.0685.5994.6689.83

KNN (correlation)84.4780.5185.9263.5686.8964.41
Improved KNN (correlation)96.1289.8396.6084.7590.2984.75

KNN (homogeneity)87.8669.4986.8959.3286.4160.17
Improved KNN (homogeneity)98.5483.0598.5485.5995.6387.29

SD represents standard deviation; bold font number indicates the best performance in each class.