Breast cancer classification from ultrasonic images based on sparse representation by exploiting redundancy

We present a Sparse Representation-based Classifier (SRC) that provides superior performance in terms of high Area Under the Receiver Operating Characteristic (ROC) Curve (AUC) in classifying benign and malignant breast lesions captured in ultrasound images. Although such a classifier was proposed for face recognition, it has been proposed in medical diagnosis from ultrasonic images in this work for the first time. The classifier is based on ℓ1-norm based sparse representation of a patient's test data in terms of linear combination of the features of the benign and malignant test lesions available in the training set. The proposed classifier introduces an index called Sparsity Rank (SR) for the classification obtained from the normalized energy of the weights as a linear combination of the global sparse representation of the ultrasound images of the training set. The performance of the classifier is further enhanced to a great extent by two ways: first, by intelligently combining the features extracted from the multiple ultrasound scan of the same mass, and the second, by using the optimal feature set obtained by a suboptimal strategy that avoids the time exhaustive brute force approach that has a combinatorial search space. With all the enhancements an AUC of 0.9802 has been achieved, when training and testing sets are chosen by leave-one-out approach from the data set.