Application of Gabor wavelet and Locality Sensitive Discriminant Analysis for automated identification of breast cancer using digitized mammogram images

摘要: Display Omitted Classification of normal, benign and malignant mammograms is proposed.Gabor wavelet coupled with Locality Sensitive Discriminant Analysis used.Achieved accuracy 98.69% for kNN classifier eight features. Breast cancer one the prime causes death in women. Early detection may help to improve survival rate a great extent. Mammography considered as most reliable methods prescreen breast cancer. However, reading by radiologists laborious, taxing, prone intra/inter observer variability errors. Computer Aided Diagnosis (CAD) helps obtain fast, consistent diagnosis. This paper presents an automated classification using digitized mammogram images. The proposed method used Gabor feature extraction (LSDA) data reduction. reduced features are ranked their F-values fed Decision Tree (DT), Linear (LDA) Quadratic (QDA), k-Nearest Neighbor (k-NN), Naive Bayes Classifier (NBC), Probabilistic Neural Network (PNN), Support Vector Machine (SVM), AdaBoost Fuzzy Sugeno (FSC) classifiers select highest performing minimum number evaluated 690 images taken from benchmarked Digital Database Screening (DDSM) dataset. Our developed has achieved mean accuracy, sensitivity, specificity 98.69%, 99.34% 98.26% respectively k-NN 10-fold cross validation. system can be employed hospitals polyclinics aid clinicians verify manual