Mammogram Analysis using League Championship Algorithm Optimized Ensembled FCRN Classifier

Abstract

An intelligent mammogram diagnosis system can be very helpful for radiologist in detecting the abnormalities earlier than typical screening techniques. This paper investigates a new classification approach for detection of breast abnormalities in digital mammograms using League Championship Algorithm Optimized Ensembled Fully Complex valued Relaxation Network (LCA-FCRN). The proposed algorithm is based on extracting curvelet fractal texture features from the mammograms and classifying the suspicious regions by applying a pattern classifier. The whole system includes steps for pre-processing, feature extraction, feature selection and classification to classify whether the given input mammogram image is normal or abnormal. The method is applied to MIAS database of 322 film mammograms. The performance of the CAD system is analysed using Receiver Operating Characteristic (ROC) curve. This curve indicates the trade-offs between sensitivity and specificity that is available from a diagnostic system, and thus describes the inherent discrimination capacity of the proposed system. The result shows that the area under the ROC curve of the proposed algorithm is 0.985 with a sensitivity of 98.1% and specificity of 92.105%. Experimental results demonstrate that the proposed method can form an effective CAD system, and achieve good classification accuracy.