Ensemble CNNs for Breast Tumor Classification

TL;DR

This paper develops an ensemble of CNN models (XceptionNet, DenseNet, EfficientNet) to improve breast tumor classification accuracy on mammographic images, achieving up to 5% performance enhancement.

Contribution

The paper introduces a novel ensemble mechanism combining multiple CNN architectures for better breast tumor classification accuracy.

Findings

Achieved 88% accuracy, 85% precision, 76% recall.

Ensemble improves performance by up to 5%.

Validated on a public dataset.

Abstract

To improve the recognition ability of computer-aided breast mass classification among mammographic images, in this work we explore the state-of-the-art classification networks to develop an ensemble mechanism. First, the regions of interest (ROIs) are obtained from the original dataset, and then three models, i.e., XceptionNet, DenseNet, and EfficientNet, are trained individually. After training, we ensemble the mechanism by summing the probabilities outputted from each network which enhances the performance up to 5%. The scheme has been validated on a public dataset and we achieved accuracy, precision, and recall 88%, 85%, and 76% respectively.