An ENAS Based Approach for Constructing Deep Learning Models for Breast Cancer Recognition from Ultrasound Images

TL;DR

This paper applies Efficient Neural Architecture Search (ENAS) to automatically design CNN models for breast cancer detection from ultrasound images, achieving high accuracy with simpler and more efficient models.

Contribution

The study introduces the use of ENAS for designing CNN architectures specifically for breast ultrasound image classification, outperforming hand-crafted models.

Findings

ENAS-generated models achieve 89.3% accuracy

Models are simpler and more efficient than traditional CNNs

ENAS is a promising approach for medical image classification

Abstract

Deep Convolutional Neural Networks (CNN) provides an "end-to-end" solution for image pattern recognition with impressive performance in many areas of application including medical imaging. Most CNN models of high performance use hand-crafted network architectures that require expertise in CNNs to utilise their potentials. In this paper, we applied the Efficient Neural Architecture Search (ENAS) method to find optimal CNN architectures for classifying breast lesions from ultrasound (US) images. Our empirical study with a dataset of 524 US images shows that the optimal models generated by using ENAS achieve an average accuracy of 89.3%, surpassing other hand-crafted alternatives. Furthermore, the models are simpler in complexity and more efficient. Our study demonstrates that the ENAS approach to CNN model design is a promising direction for classifying ultrasound images of breast lesions.