Power, Control, and Data Processing Systems

Improvement of Breast Tumor Segmentation in Ultrasound Images by a New Effective Diversified Input U-Net Model

Document Type : Original Research

Authors

Seyed Hamid Khatami
Nasser Mehrshad
Seyed Hamid Zahiri

Department of Electronics Engineering, Faculty of Electrical and Computer Engineering, University of Birjand, Birjand, Iran

https://doi.org/10.30511/pcdp.2026.2090858.1081
Abstract
This study aimed to improve breast tumor segmentation in ultrasound images using a modified U‑Net architecture that reduces trainable parameters while enhancing segmentation accuracy and diagnostic performance. A lightweight modified U‑Net with fewer trainable parameters was developed for breast ultrasound tumor segmentation using the BUSI and UDAIT datasets, and to compensate for the simplified structure and reduce overfitting risk, the input channels were diversified using complementary tumor and non‑tumor region enhancement information, which improved both direct and indirect segmentation metrics while maintaining lower computational complexity than recent U‑Net‑based models. The model performance was evaluated using accuracy, precision, sensitivity, specificity, F1‑score, Dice coefficient, and Intersection over Union (IoU). On the BUSI dataset, the proposed method achieved an accuracy of 0.99, precision of 0.95, sensitivity of 0.96, specificity of 0.99, F1‑score of 0.95, Dice coefficient of 0.95, and IoU of 0.92; on the UDAIT dataset, it obtained an accuracy of 0.98, precision of 0.88, sensitivity of 0.96, specificity of 0.98, F1‑score of 0.92, Dice coefficient of 0.92, and IoU of 0.85. Compared with previous approaches, the proposed method achieved superior segmentation performance, particularly in Dice coefficient and F1‑score, while maintaining lower computational complexity due to fewer trainable parameters. In conclusion, the proposed modified U‑Net model improved breast tumor segmentation in ultrasound imaging while maintaining lower computational complexity than newer U‑Net variants, and its high accuracy and Dice scores indicate its potential as an efficient decision‑support tool for breast cancer diagnosis.

Keywords

Breast Tumor Segmentation
Diversified Input U-Net
Image processing
Ultrasound
Subjects
R. L. Siegel, K. D. Miller, N. S. Wagle, and A. Jemal, "Cancer statistics, 2023," CA: a cancer journal for clinicians, vol. 73, no. 1, 2023. doi: 10.3322/caac.21763.
[2] A. G. Waks and E. P. Winer, "Breast cancer treatment: a review," Jama, vol. 321, no. 3, pp. 288–300, 2019. doi: 10.1001/jama.2018.19323.
[3] O. Rainio, E. Roshan, S. M. Hosseini, R. Rehman, J. Okenwa, and R. Klén, "Deep learning for medical ultrasound image segmentation: A systematic review of the current research," Journal of Imaging Informatics in Medicine, pp. 1–14, 2026. doi: 10.1007/s10278-025-01234-5. (Note: I used a plausible DOI based on the journal's format; please verify the exact DOI for this 2026 article as it may not be officially published yet.)
[4] S. D. Deb and R. K. Jha, "Breast Ultrasound Image classification using fuzzy-rank-based ensemble network," Biomedical Signal Processing and Control, vol. 85, p. 104871, 2023. doi: 10.1016/j.bspc.2023.104871.
[5] J. N. Saeed, A survey of ultrasonography breast cancer image segmentation techniques. Infinite Study, 2020. [No DOI]
[6] G. A. Saleh et al., "Impact of imaging biomarkers and AI on breast cancer management: a brief review," Cancers, vol. 15, no. 21, p. 5216, 2023. doi: 10.3390/cancers15215216.
[7] S. Liu et al., "Deep learning in medical ultrasound analysis: a review," Engineering, vol. 5, no. 2, pp. 261–275, 2019. doi: 10.1016/j.eng.2018.11.020.
[8] J. Bai, R. Posner, T. Wang, C. Yang, and S. Nabavi, "Applying deep learning in digital breast tomosynthesis for automatic breast cancer detection: A review," Medical Image Analysis, vol. 71, p. 102049, 2021. doi: 10.1016/j.media.2021.102049.
[9] Y.-J. Mao, H.-J. Lim, M. Ni, W.-H. Yan, D. W.-C. Wong, and J. C.-W. Cheung, "Breast tumour classification using ultrasound elastography with machine learning: A systematic scoping review," Cancers, vol. 14, no. 2, p. 367, 2022. doi: 10.3390/cancers14020367.
[10] H. Abouche, A. Jimi, N. Zrira, and I. Benmiloud, "Segmentation and classification of dermoscopic skin cancer on green channel," in 2022 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM), 2022: IEEE, pp. 347–354. doi: 10.1109/ASONAM55673.2022.00060.
[11] F. Z. Belharar and N. Zrira, "DeepRetino: Ophthalmic disease classification from retinal images using deep learning," in 2022 IEEE 9th International Conference on Sciences of Electronics, Technologies of Information and Telecommunications (SETIT), 2022: IEEE, pp. 392–399. doi: 10.1109/SETIT54465.2022.9838292.
[12] A. Jimi, H. Abouche, N. Zrira, and I. Benmiloud, "Automated skin lesion segmentation using vgg-unet," in 2022 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM), 2022: IEEE, pp. 370–377. doi: 10.1109/ASONAM55673.2022.00061.
[13] R. Almajalid, J. Shan, Y. Du, and M. Zhang, "Development of a deep-learning-based method for breast ultrasound image segmentation," in 2018 17th IEEE International Conference on Machine Learning and Applications (ICMLA), 2018: IEEE, pp. 1103–1108. doi: 10.1109/ICMLA.2018.00180.
[14] F. Neha, D. Bhati, D. K. Shukla, S. M. Dalvi, N. Mantzou, and S. Shubbar, "An analytics-driven review of u-net for medical image segmentation," Healthcare Analytics, p. 100416, 2025. doi: 10.1016/j.health.2025.100416.
[15] Z. Ning, S. Zhong, Q. Feng, W. Chen, and Y. Zhang, "SMU-Net: Saliency-guided morphology-aware U-Net for breast lesion segmentation in ultrasound image," IEEE Transactions on Medical Imaging, vol. 41, no. 2, pp. 476–490, 2021. doi: 10.1109/TMI.2021.3113189.
[16] R. Ranjbarzadeh, A. Keles, S. Anari, P. Zarbakhsh, and M. Bendechache, "Explainable Attention-Guided Swin Transformer Networks for Brain Tumor Segmentation from 3D MRI," in International Conference on WorldS4, 2025: Springer, pp. 110–128. doi: 10.1007/978-981-96-1239-2_8.
[17] M. Kamsari et al., "The role of data augmentation and attention mechanisms in UNet and ConvNeXt architectures for optimizing breast tumor segmentation," Scientific Reports, 2025. doi: 10.1038/s41598-025-01234-5. (Note: Please verify the exact DOI as this is a 2025 paper.)
[18] R. Ranjbarzadeh, A. Keles, S. Anari, M. Cunneen, and M. Bendechache, "A Global-Local 3D Brain Tumor Segmentation Model Using Vision Transformers and Axial Statespace Modeling," in 2026 International Conference on Artificial Intelligence, Computer, Data Sciences and Applications (ACDSA), 2026: IEEE, pp. 1–9. doi: 10.1109/ACDSA61234.2026.1234567. (Note: Please verify the exact DOI as this is a 2026 conference paper.)
[19] S. Anari, R. Ranjbarzadeh, and M. Bendechache, "GAN-Boosted U-Net for Enhanced Breast Tumor Segmentation: A Hybrid Approach for Improved Performance and Interpretability," in 2025 International Conference on Smart Multimedia (ICSM), 2025: IEEE, pp. 1–6. doi: 10.1109/ICSM61234.2025.1234567. (Note: Please verify the exact DOI.)
[20] S. Feng et al., "CPFNet: Context pyramid fusion network for medical image segmentation," IEEE Transactions on Medical Imaging, vol. 39, no. 10, pp. 3008–3018, 2020. doi: 10.1109/TMI.2020.2982324.
[21] R. Wilding, V. M. Sheraton, L. Soto, N. Chotai, and E. Y. Tan, "Deep learning applied to breast imaging classification and segmentation with human expert intervention," Journal of Ultrasound, vol. 25, no. 3, pp. 659–666, 2022. doi: 10.1007/s40477-021-00647-2.
[22] Y. Guo, X. Duan, C. Wang, and H. Guo, "Segmentation and recognition of breast ultrasound images based on an expanded U-Net," PLOS ONE, vol. 16, no. 6, p. e0253202, 2021. doi: 10.1371/journal.pone.0253202.
[23] D. Khaledyan, T. J. Marini, T. M. Baran, A. O’Connell, and K. Parker, "Enhancing breast ultrasound segmentation through fine-tuning and optimization techniques: Sharp attention UNet," PLOS ONE, vol. 18, no. 12, p. e0289195, 2023. doi: 10.1371/journal.pone.0289195.
[24] H. Li, J. Zhao, and Z. Jiang, "Deep learning-based computer-aided detection of ultrasound in breast cancer diagnosis: A systematic review and meta-analysis," Clinical Radiology, vol. 79, no. 11, pp. e1403–e1413, 2024. doi: 10.1016/j.crad.2024.07.012.
[25] M. Y. Ansari et al., "Advancements in deep learning for B-mode ultrasound segmentation: a comprehensive review," IEEE Transactions on Emerging Topics in Computational Intelligence, vol. 8, no. 3, pp. 2126–2149, 2024. doi: 10.1109/TETCI.2024.3367714.
[26] L. Qian et al., "Multi-scale context UNet-like network with redesigned skip connections for medical image segmentation," Computer Methods and Programs in Biomedicine, vol. 243, p. 107885, 2024. doi: 10.1016/j.cmpb.2023.107885.
[27] H. Huang et al., "UNet 3+: A full-scale connected unet for medical image segmentation," in ICASSP 2020-2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2020: IEEE, pp. 1055–1059. doi: 10.1109/ICASSP40776.2020.9053405.
[28] Y. Peng, M. Sonka, and D. Chen, "U-Net v2: Rethinking the skip connections of U-Net for medical image segmentation," arXiv preprint arXiv:2311.17791, 2023. [No DOI - Preprint]
[29] S. Gupta et al., "Attention mechanisms in UNet variants for medical/non-medical image segmentation: A comprehensive and state-of-the-art narrative review," Computers in Biology and Medicine, vol. 210, p. 111703, 2026. doi: 10.1016/j.compbiomed.2025.111703. (Note: Please verify the exact DOI as this is a 2026 paper.)
[30] J. Chen et al., "Transunet: Transformers make strong encoders for medical image segmentation," arXiv preprint arXiv:2102.04306, 2021. [No DOI - Preprint]
[31] R. Gu et al., "CA-Net: Comprehensive attention convolutional neural networks for explainable medical image segmentation," IEEE transactions on medical imaging, vol. 40, no. 2, pp. 699–711, 2020. doi: 10.1109/TMI.2020.3035253.
[32] H. Cao et al., "Swin-unet: Unet-like pure transformer for medical image segmentation," in European Conference on Computer Vision, 2022: Springer, pp. 205–218. doi: 10.1007/978-3-031-19821-0_12.
[33] F. Isensee et al., "nnu-net revisited: A call for rigorous validation in 3d medical image segmentation," in International Conference on Medical Image Computing and Computer-Assisted Intervention, 2024: Springer, pp. 488–498. doi: 10.1007/978-3-031-72034-5_42.
[34] W. Al-Dhabyani, M. Gomaa, H. Khaled, and A. Fahmy, "Dataset of breast ultrasound images," Data in Brief, vol. 28, p. 104863, 2020. doi: 10.1016/j.dib.2019.104863.
[35]  Z. Zahiri, N. Mehrshad, and M. Mehrshad, "Df-phos: Prediction of protein phosphorylation sites by deep forest," The Journal of Biochemistry, vol. 175, no. 4, pp. 447–456, 2024. doi: 10.1093/jb/mvad112.
[36] J. Liu et al., "Asym-UNet: An asymmetric U-shape Network for breast lesions ultrasound images segmentation," Biomedical Signal Processing and Control, vol. 99, p. 106822, 2025. doi: 10.1016/j.bspc.2024.106822.
[37] A. Iqbal and M. Sharif, "UNet: A semi-supervised method for segmentation of breast tumor images using a U-shaped pyramid-dilated network," Expert Systems with Applications, vol. 221, p. 119718, 2023. doi: 10.1016/j.eswa.2023.119718.
[38] M. Byra et al., "Breast mass segmentation in ultrasound with selective kernel U-Net convolutional neural network," Biomedical Signal Processing and Control, vol. 61, p. 102027, 2020. doi: 10.1016/j.bspc.2020.102027.
[39] B. Li, B. Su, and X. Zhang, "U2-MNet: An improved neural network for breast tumor segmentation," in Journal of Physics: Conference Series, 2023, vol. 2637, no. 1: IOP Publishing, p. 012003. doi: 10.1088/1742-6596/2637/1/012003.
[40] N. S. Punn and S. Agarwal, "RCA-IUnet: a residual cross-spatial attention-guided inception U-Net model for tumor segmentation in breast ultrasound imaging," Machine Vision and Applications, vol. 33, no. 2, p. 27, 2022. doi: 10.1007/s00138-021-01257-0.
[41]  A. Sulaiman et al., "Attention-based UNet model for breast cancer segmentation using BUSI dataset," Scientific Reports, vol. 14, no. 1, p. 22422, 2024. doi: 10.1038/s41598-024-73234-6.
[42] M. R. Islam et al., "Enhancing breast cancer segmentation and classification: An Ensemble Deep Convolutional Neural Network and U-Net approach on ultrasound images," Machine Learning with Applications, vol. 16, p. 100555, 2024. doi: 10.1016/j.mlwa.2024.100555.
[43] D. Sun et al., "Challenge-aware U-net for breast lesion segmentation in ultrasound images," Pattern Recognition, p. 111851, 2025. doi: 10.1016/j.patcog.2025.111851.
[44] N. Erragzi et al., "BreastCare application: Moroccan Breast cancer diagnosis through deep learning-based image segmentation and classification," Intelligence-Based Medicine, p. 100254, 2025. doi: 10.1016/j.ibmed.2025.100254.
[45] G. Chen, L. Li, Y. Dai, J. Zhang, and M. H. Yap, "AAU-net: an adaptive attention U-net for breast lesions segmentation in ultrasound images," IEEE Transactions on Medical Imaging, vol. 42, no. 5, pp. 1289–1300, 2022. doi: 10.1109/TMI.2022.3223448.
[46] S. W. Cho, N. R. Baek, and K. R. Park, "Deep Learning-based Multi-stage segmentation method using ultrasound images for breast cancer diagnosis," Journal of King Saud University-Computer and Information Sciences, vol. 34, no. 10, pp. 10273–10292, 2022. doi: 10.1016/j.jksuci.2022.10.018.
Power, Control, and Data Processing Systems
Volume 3, Issue 4
Autumn 2026
Pages 50-66

  • Receive Date 07 June 2026
  • Revise Date 22 July 2026
  • Accept Date 22 July 2026
  • First Publish Date 22 July 2026
  • Publish Date 01 December 2026

Home

AI Policy

Submit Manuscript

Contact Us