Convolutional Neural Networks for Breast Cancer Screening: Transfer Learning with Exponential Decay

TL;DR

This paper presents a CNN-based CAD system for breast cancer screening that leverages transfer learning and fine-tuning to improve classification accuracy with limited data.

Contribution

It introduces a transfer learning approach with optimized fine-tuning for CNNs in breast lesion classification, enhancing performance with scarce data.

Findings

Transfer learning improves classification accuracy.

Proper layer fine-tuning enhances feature specificity.

Proposed method outperforms existing models on the dataset.

Abstract

In this paper, we propose a Computer Assisted Diagnosis (CAD) system based on a deep Convolutional Neural Network (CNN) model, to build an end-to-end learning process that classifies breast mass lesions. We investigate the impact that has transfer learning when large data is scarce, and explore the proper way to fine-tune the layers to learn features that are more specific to the new data. The proposed approach showed better performance compared to other proposals that classified the same dataset.