Breast Cancer Detection Using Extreme Learning Machine Based on Feature Fusion With CNN Deep Features

Tags: , , ,

Aim:

To classify benign and malignant breast masses using K Nearest Neighbor based on Feature Fusion with CNN Deep Features.

Existing System:

K Nearest Neighbor has better classification effect on multi-dimensional features than other classifiers including Logistic regression, decision tree, etc.,based on our previous research. Thus, we use KNN to classify the extracted breast mass features. Therefore, in this paper, we propose a novel diagnosis method that merges several deep features.

Proposed System:

We propose a mass detection method based on CNN deep features and unsupervised K Nearest Neighbor (KNN) clustering. Second, we build a feature set fusing deep features, morphological features, texture features, and density features. Third, an KNN classifier is developed using the fused feature set to classify benign and malignant breast masses. Extensive experiments demonstrate the accuracy and efficiency of our proposed mass detection and breast cancer classification method.