Available online 13 March 2023, 100774

We employed transfer learning to get image representations using pre-trained CNN models.

We proposed an adaptive backbone model selection algorithm to obtain the best CNN model.

an ELM was trained by the representations and the labels as the classifier in the proposed BCDNet.

A novel bat algorithm with chaotic maps (BACM) was introduced to optimize the parameters.

Our BCDNet outperformed state-of-the-art breast cancer detection methods in terms of accuracy.

Breast cancer is a common but deadly disease among women. Medical imaging is an effective method to diagnose breast cancer, but manual image screening is time-consuming. In this study, a novel computer-aided diagnosis system for breast cancer detection called BCDNet is proposed.

We leverage pre-trained convolutional neural networks (CNNs) for representation learning and propose an adaptive backbone selection algorithm to obtain the best CNN model. An extreme learning machine serves as the classifier in the BCDNet, and a bat algorithm with chaotic maps is put forward to further optimize the parameters in the classifiers. A public ultrasound image dataset is used in the experiments based on 5-fold cross-validation.

Simulation results suggest that our BCDNet outperforms several state-of-the-art breast cancer detection methods in terms of accuracy.

The proposed BCDNet is a useful auxiliary tool that can be applied in clinical screening for breast cancer.

computer-aided diagnosis

ultrasound image

deep learning

extreme learning machine

swarm intelligence

© 2023 AGBM. Published by Elsevier Masson SAS.