Breast cancer is the most prevalent and fatal cancer among women globally, with breast ultrasound imaging often being the primary diagnostic method in clinical examinations. However, challenges such as difficulty in acquisition, limited annotation, and inconsistencies in data feature distribution result in low accuracy for both manual and traditional network model-assisted diagnoses. To address these issues, this study proposed a domain adaptation network based on adversarial-learned loss and deep correlation alignment (ALCOR_DA). Building upon the foundational framework of dynamic adversarial adaptation networks (DAANs), the proposed model introduces a novel loss function, the adversarial-learned loss. By leveraging the adversarial interaction between the generator and domain discriminator within the framework, a noise-confusion matrix is generated to refine the pseudo-labels produced by the classifier for target domain data. This process reduces the discrepancy between pseudo-labels and true labels, thereby improving the classification accuracy for unlabeled target domain data. Additionally, the model incorporates an adaptive layer and employs the deep correlation alignment algorithm to measure and align the feature distributions between the source and target domains. This alignment enhances the generalization capability of the model across different datasets. On public datasets, ALCOR_DA achieved an accuracy of 86.26%, representing a 4.65% improvement over the traditional DAAN model, underscoring its effectiveness in practical diagnostic scenarios.

1. [1] H. Sung et al., “Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries,” CA: A Cancer J. for Clinicians, Vol.71, No.3, pp. 209-249, 2021. https://doi.org/10.3322/caac.21660
2. [2] A. B. M. A. Hossain, J. K. Nisha, and F. Johora, “Breast cancer classification from ultrasound images using VGG16 model based transfer learning,” Int. J. of Image, Graphics and Signal Processing, Vol.15, No.1, pp. 12-22, 2023. https://doi.org/10.5815/ijigsp.2023.01.02
3. [3] W. Islam et al., “Improving performance of breast lesion classification using a ResNet50 model optimized with a novel attention mechanism,” Tomography, Vol.8, No.5, pp. 2411-2425, 2022. https://doi.org/10.3390/tomography8050200
4. [4] L. Fan, X. Gong, and Y. Guo, “General multiscenario ultrasound image tumor diagnosis method based on unsupervised domain adaptation,” Ultrasound in Medicine and Biology, Vol.49, No.10, pp. 2291-2301, 2023. https://doi.org/10.1016/j.ultrasmedbio.2023.06.015
5. [5] S. Muksimova, S. Umirzakova, K. Shoraimov, J. Baltayev, and Y.-I. Cho, “Novelty classification model use in reinforcement learning for cervical cancer,” Cancers, Vol.16, No.22, Article No.3782, 2024. https://doi.org/10.3390/cancers16223782
6. [6] J. Ru, Z. Zhu, and J. Shi, “Spatial and geometric learning for classification of breast tumors from multi-center ultrasound images: A hybrid learning approach,” BMC Medical Imaging, Vol.24, Article No.133, 2024. https://doi.org/10.1186/s12880-024-01307-3
7. [7] B. Xu, K. Wu, Y. Wu, J. He, and C. Chen, “Dynamic adversarial domain adaptation based on multikernel maximum mean discrepancy for breast ultrasound image classification,” Expert Systems with Applications, Vol.207, Article No.117978, 2022. https://doi.org/10.1016/j.eswa.2022.117978
8. [8] N. M. Ud Din, R. A. Dar, M. Rasool, and A. Assad, “Breast cancer detection using deep learning: Datasets, methods, and challenges ahead,” Computers in Biology and Medicine, Vol.149, Article No.106073, 2022. https://doi.org/10.1016/j.compbiomed.2022.106073
9. [9] C. Yu, J. Wang, Y. Chen, and M. Huang, “Transfer learning with dynamic adversarial adaptation network,” 2019 IEEE Int. Conf. on Data Mining, pp. 778-786, 2019. https://doi.org/10.1109/ICDM.2019.00088
10. [10] M. Chen, S. Zhao, H. Liu, and D. Cai, “Adversarial-learned loss for domain adaptation,” Proc. of the AAAI Conf. on Artificial Intelligence, Vol.34, No.4, pp. 3521-3528, 2020. https://doi.org/10.1609/aaai.v34i04.5757
11. [11] B. Sun and K. Saenko, “Deep CORAL: Correlation alignment for deep domain adaptation,” Computer Vision: ECCV 2016 Workshops, Part 3, pp. 443-450, 2016. https://doi.org/10.1007/978-3-319-49409-8_35
12. [12] M. Song and Y. Kim, “Unsupervised learning method via triple reconstruction for the classification of ultrasound breast lesions,” Biomedical Signal Processing and Control, Vol.77, Article No.103782, 2022. https://doi.org/10.1016/j.bspc.2022.103782
13. [13] O. Tmenova, Y. Velikova, M. Saleh, and N. Navab, “Deep spectral methods for unsupervised ultrasound image interpretation,” Proc. of Medical Image Computing and Computer Assisted Intervention, Part 11, pp. 200-210, 2024. https://doi.org/10.1007/978-3-031-72120-5_19
14. [14] E. Tzeng, J. Hoffman, N. Zhang, K. Saenko, and T. Darrell, “Deep domain confusion: Maximizing for domain invariance,” arXiv:1412.3474, 2014. https://doi.org/10.48550/arXiv.1412.3474
15. [15] K. M. Borgwardt et al., “Integrating structured biological data by kernel maximum mean discrepancy,” Bioinformatics, Vol.22, No.14, pp. e49-e57, 2006. https://doi.org/10.1093/bioinformatics/btl242
16. [16] M. Long, Y. Cao, J. Wang, and M. Jordan, “Learning transferable features with deep adaptation networks,” Proc. of the 32nd Int. Conf. on Machine Learning, pp. 97-105, 2015.
17. [17] Y. Ganin et al., “Domain-adversarial training of neural networks,” J. of Machine Learning Research, Vol.17, Article No.59, 2016.
18. [18] T.-R. Wei et al., “Enhancing AI diagnostics: Autonomous lesion masking via semi-supervised deep learning,” arXiv:2404.12450, 2024. https://doi.org/10.48550/arXiv.2404.12450
19. [19] B. Sun, J. Feng, and K. Saenko, “Return of frustratingly easy domain adaptation,” Proc. of the AAAI Conf. on Artificial Intelligence, Vol.30, No.1, pp. 2058-2065, 2016. https://doi.org/10.1609/aaai.v30i1.10306
20. [20] M. H. Yap et al., “Automated breast ultrasound lesions detection using convolutional neural networks,” IEEE J. of Biomedical and Health Informatics, Vol.22, No.4, pp. 1218-1226, 2018. https://doi.org/10.1109/JBHI.2017.2731873