Call For Paper - Upcoming Conferences
Research Article | Open Access | Download PDF
Volume 13 | Issue 4 | Year 2026 | Article Id. IJEEE-V13I4P109 | DOI : https://doi.org/10.14445/23488379/IJEEE-V13I4P109

An Efficient DL Based Segmentation Technique for Breast Ultrasound Images

Priyanshu Tripathi, Rajeshwar Dass, Jyotsna Sen

Received Revised Accepted Published
18 Jan 2026 27 Feb 2026 26 Mar 2026 30 Apr 2026

Citation :

Priyanshu Tripathi, Rajeshwar Dass, Jyotsna Sen, "An Efficient DL Based Segmentation Technique for Breast Ultrasound Images," International Journal of Electrical and Electronics Engineering, vol. 13, no. 4, pp. 119-132, 2026. Crossref, https://doi.org/10.14445/23488379/IJEEE-V13I4P109

Abstract

This paper presents an efficient and novel DL-based segmentation technique for Breast Ultrasonographic (BUS) Images. In the first step, BUS images are despeckled to get enhanced lesion texture and visibility. Further, an encoder-decoder-based segmentation algorithm is employed for the accurate detection of the boundary and textual information of the breast tissue. Various pre-existing despeckling filters are categorized into basic, smoothing, and edge-preserving filters, and their performance is evaluated using SEPI metrics. Finally, the efficient despeckling filters are fused to enhance performance. The fusion of the Bilateral Filter and Anisotropic Diffusion Filter (BiF-ADF) yields optimal performance according to the SEPI metrics. After despeckling, the filtered images are fed into a semantic segmentation module to segment lesions in the BUS images. In the segmentation module, various state-of-the-art DL-based algorithms are implemented using transfer learning, and their performance is evaluated based on the mean IOU. It is observed that the ResNet-50 model yielded a higher value. To further enhance the segmentation performance, an efficient technique, Clipped-ResNet50, is proposed, yielding the highest mean IOU values of 0.9089 and 0.9060, and accuracy of 0.9379 and 0.9387 for the BUSI and PGI Rohtak datasets, respectively. Further, Clipped-ResNet50 is also trained on despeckled, i.e., enhanced images, showing a notable improvement in segmentation outcomes. When applying the model to despeckled images, the mean IOU and accuracy increased from 90.89% to 93.05% and 90.60% to 92.89% for the BUSI and PGI Rohtak datasets, respectively, highlighting the significant advantage of using despeckled breast ultrasound images for segmentation tasks.

Keywords

Breast Ultrasonographic (BUS) Images, Speckle Noise, Despeckling Filters, SEPI metrics, Image Fusion, Semantic Segmentation, Mean IOU.

References

  1. Kathryn P. Trayes, and Sarah E.H. Cokenakes, “Breast Cancer Treatment,” American Family Physician, vol. 104, no. 2, pp. 171-178, 2021.
    [    Google Scholar] [Publisher Link]
  2. Melina Arnold et al., “Current and Future Burden of Breast Cancer: Global Statistics for 2020 and 2040,” The Breast, vol. 66, pp. 15-23, 2022.
    [    CrossRef] [Google Scholar] [Publisher Link]
  3. Mohammad Madani, Mohammad Mahdi Behzadi, and Sheida Nabavi, “The Role of Deep Learning in Advancing Breast Cancer Detection using Different Imaging Modalities: A Systematic Review,” Cancers, vol. 14, no. 21, pp. 1-36, 2022.
    [    CrossRef] [Google Scholar] [Publisher Link]
  4. Athar Shams Rana, Javairia Rafique, and Hira Riffat, Advances in Breast Ultrasound Imaging: Enhancing Diagnostic Precision and Clinical Utility, Latest Research on Breast Cancer, 2024.
    [CrossRef] [Google Scholar] [Publisher Link]
  5. Wilfrido Gómez Flores et al., “Breast Ultrasound Despeckling using Anisotropic Diffusion Guided by Texture Descriptors,” Ultrasound in Medicine & Biology, vol. 40, no. 11, pp. 2609-2621, 2014.
    [    CrossRef] [Google Scholar] [Publisher Link]
  6. Iman Elyasi et al., “Speckle Reduction in Breast Cancer Ultrasound Images by using Homogeneity Modified Bayes Shrink,” Measurement, vol. 91, pp. 55-65, 2016.
    [    CrossRef] [Google Scholar] [Publisher Link]
  7. Kriti, Jitendra Virmani, and Ravinder Agarwal, “Assessment of Despeckle Filtering Algorithms for Segmentation of Breast Tumours from Ultrasound Images,” Biocybernetics and Biomedical Engineering, vol. 39, no. 1, pp. 100-121, 2019.
    [    CrossRef] [Google Scholar] [Publisher Link]
  8. Hyunho, and Jechang Jeong, “Despeckling Algorithm for Removing Speckle Noise from Ultrasound Images,” Symmetry, vol. 12, no. 6, pp. 1-26, 2020.
    [    CrossRef] [Google Scholar] [Publisher Link]
  9. Gelan Ayana et al., “De-Speckling Breast Cancer Ultrasound Images using a Rotationally Invariant Block Matching Based Non-Local Means (RIBM-NLM) Method,” Diagnostics, vol. 12, no. 4, pp. 1-17, 2022.
    [    CrossRef] [Google Scholar] [Publisher Link]
  10. Niranjan Yadav, Rajeshwar Dass, and Jitendra Virmani, “Despeckling Filters Applied to Thyroid Ultrasound Images: A Comparative Analysis,” Multimedia Tools and Applications, vol. 81, pp. 8905-8937, 2022.
    [    CrossRef] [Google Scholar] [Publisher Link]
  11. Tong Ying et al., “Breast Ultrasound Image Despeckling using Multi-Filtering DFrFT and Adaptive Fast BM3D,” Computer Methods and Programs in Biomedicine, vol. 246, 2024.
    [    CrossRef] [Google Scholar] [Publisher Link]
  12. Dongkyu Jung et al., “Unsupervised Speckle Noise Reduction Technique for Clinical Ultrasound Imaging,” Ultrasonography, vol. 43, no. 5, pp. 327-344, 2024.
    [    CrossRef] [Google Scholar] [Publisher Link]
  13. Xiangyu Li et al., “A Detail-Preserving Ultrasound Speckle Reduction Method Based on Complementary Information,” Ultrasound in Medicine & Biology, vol. 52, no. 1, pp. 108-122, 2026.
    [    CrossRef] [Google Scholar] [Publisher Link]
  14. Mostafa Khalifa, Hanaa M. Hamza, and Khalid M. Hosny, “De-speckling of Medical Ultrasound Image using Metric-Optimized Knowledge Distillation,” Scientific Reports, vol. 15, pp. 1-21, 2025.
    [    CrossRef] [Google Scholar] [Publisher Link]
  15. M.S. Gokmen et al., “An Effective Image Despeckling and Reconstruction Approach using U-Net based Model and Comparative Analysis,” Scientific Reports, vol. 15, pp. 1-21, 2025.
    [    CrossRef] [Google Scholar] [Publisher Link]
  16. Nilima Patil, M.M. Deshpande, and V.N. Pawar, “A Novel Deep Learning Approach for Speckle Denoising Using Hyperparameter Tuning,” Journal of Image and Graphics, vol. 13, no. 3, pp. 267-274, 2025.
    [CrossRef] [Google Scholar] [Publisher Link]
  17. Vijaya S. Patil et al., “Despeckeling Method for Ultrasound Thyroid Nodules Using Innovative Wiener Filter,” Proceedings of the 3rd International Conference on Futuristic Technology, vol. 2, pp. 885-889, 2025.
    [CrossRef] [Google Scholar] [Publisher Link]
  18. Walid Al-Dhabyani, Dataset of Breast Ultrasound Images, Data in Brief, vol. 28, pp. 1-5, 2020.
    [    CrossRef] [Google Scholar] [Publisher Link]
  19. Fatima M. Osman, and Moi Hoon Yap, “The Effect of Filtering Algorithms for Breast Ultrasound Lesions Segmentation,” Informatics in Medicine Unlocked, vol. 12, pp. 14-20, 2018.
    [    CrossRef] [Google Scholar] [Publisher Link]
  20. Ju Zhang, Chen Wang, and Yun Cheng, “Comparison of Despeckle Filters for Breast Ultrasound Images,” Circuits, Systems, and Signal Processing, vol. 34, pp. 185-208, 2015.
    [    CrossRef] [Google Scholar] [Publisher Link]
  21. Jong-Sen Lee, “Speckle Analysis and Smoothing of Synthetic Aperture Radar Images,” Computer Graphics and Image Processing, vol. 17, no. 1, pp. 24-32, 1981.
    [    CrossRef] [Google Scholar] [Publisher Link]
  22. Darwin T. Kuan et al., “Adaptive Noise Smoothing Filter for Images with Signal-Dependent Noise,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 7, no. 2, pp. 165-177, 1985.
    [    CrossRef] [Google Scholar] [Publisher Link]
  23. Victor S. Frost et al., “A Model for Radar Images and its Application to Adaptive Digital Filtering of Multiplicative Noise,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 4, no. 2, pp. 157-166, 1982.
    [    CrossRef] [Google Scholar] [Publisher Link]
  24. P. Perona, and J. Malik, “Scale-Space and Edge Detection Using Anisotropic Diffusion,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 12, no. 7, pp. 629-639, 1990.
    [    CrossRef] [Google Scholar] [Publisher Link]
  25. Irraivan Elamvazuthi, Muhammad Luqman Bin Muhd Zain, and K.M. Begam, “Despeckling of Ultrasound Images of Bone Fracture using Multiple Filtering Algorithms,” Mathematical and Computer Modelling, vol. 57, no. 1-2, pp. 152-168, 2013.
    [    CrossRef] [Google Scholar] [Publisher Link]
  26. Joachim Weickert, “Coherence-Enhancing Diffusion Filtering,” International Journal of Computer Vision, vol. 31, pp. 111-127, 1999.
    [    CrossRef] [Google Scholar] [Publisher Link]
  27. Sean Finn, Martin Glavin, and Edward Jones, “Echocardiographic Speckle Reduction Comparison,” IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control, vol. 58, no. 1, pp. 82-101, 2011.
    [    CrossRef] [Google Scholar] [Publisher Link]
  28. Rajeshwar Dass, and Niranjan Yadav, “Image Quality Assessment Parameters for Despeckling Filters,” Procedia Computer Science, vol. 167, pp. 2382-2392, 2020.
    [    CrossRef] [Google Scholar] [Publisher Link]
  29. S. Senthil Kumar, and S. Muttan, “PCA-based Image Fusion,” Algorithms and Technologies for Multispectral, Hyperspectral, and Ultraspectral Imagery XII, vol. 6233, 2006.
    [    CrossRef] [Google Scholar] [Publisher Link]
  30. Samuel E. Johnny et al., “Prompt-Free SAM-Based Multi-Task Framework for Breast Ultrasound Lesion Segmentation and Classification,” arXiv preprint, pp. 1-6, 2026.
    [    CrossRef] [Google Scholar] [Publisher Link]
  31. Yajuan Li et al., “Deep Learning Based on Ultrasound Images Differentiates Parotid Gland Pleomorphic Adenomas and Warthin Tumors,” Ultrasonic Imaging, vol. 47, no. 3-4, 2025.
    [    CrossRef] [Google Scholar] [Publisher Link]
  32. Pierangela Bruno, Megan Macrì, and Carmine Dodaro, “A Dual-Stage Deep Learning Framework for Breast Ultrasound Image Segmentation and Classification,” Journal of Medical Systems, vol. 49, pp. 1-11, 2025.
    [    CrossRef] [Google Scholar] [Publisher Link]
  33. Mohammad Amanour Rahman, “HyFormer-Net: A Synergistic CNN-Transformer with Interpretable Multi-Scale Fusion for Breast Lesion Segmentation and Classification in Ultrasound Images,” arXiv preprint, pp. 1-16, 2025.
    [    CrossRef] [Google Scholar] [Publisher Link]
  34. Lal Omega Boro, and Gypsy Nandi, “CBAM-RIUnet: Breast Tumor Segmentation with Enhanced Breast Ultrasound and Test-Time Augmentation,” Ultrasonic Imaging, vol. 47, no. 1, pp. 24-36, 2024.
    [    CrossRef] [Google Scholar] [Publisher Link]
  35. Golla Madhu et al., “UCapsNet: A Two-Stage Deep Learning Model Using U-Net and Capsule Network for Breast Cancer Segmentation and Classification in Ultrasound Imaging,” Cancers, vol. 16, no. 22, pp. 1-20, 2024.
    [    CrossRef] [Google Scholar] [Publisher Link]
  36. Payel Pramanik et al., “DAU-Net: Dual Attention-aided U-Net for Segmenting Tumor in Breast Ultrasound Images,” Plos one, vol. 19, no. 5, pp. 1-20, 2024.
    [    CrossRef] [Google Scholar] [Publisher Link]
  37. Saba Hesaraki, Morteza Akbari, and Ramin Mousa, “UNet++ and Lstm Combined Approach for Breast Ultrasound Image Segmentation,” arXiv preprint, pp. 1-16, 2024.
    [    CrossRef] [Google Scholar] [Publisher Link]
  38. Yuchao Lyu et al., “AMS-PAN: Breast Ultrasound Image Segmentation Model Combining Attention Mechanism and Multi-Scale Features,” Biomedical Signal Processing and Control, vol. 81, pp. 1-13, 2023.
    [    CrossRef] [Google Scholar] [Publisher Link]
  39. Haonan Yang, and Dapeng Yang, “CSwin-PNet : A CNN-Swin Transformer Combined Pyramid Network for Breast Lesion Segmentation in Ultrasound Images,” Expert Systems with Applications, vol. 213, 2023.
    [    CrossRef] [Google Scholar] [Publisher Link]
  40. Tianyu Zhao, and Hang Dai, “Breast Tumor Ultrasound Image Segmentation Method Based on Improved Residual U-Net Network,” Computational Intelligence and Neuroscience, vol. 2022, no. 1, pp. 1-9, 2022.
    [    CrossRef] [Google Scholar] [Publisher Link]
  41. Jaeil Kim et al., “Weakly ‑ Supervised Deep Learning for Ultrasound Diagnosis of Breast Cancer,” Scientific Reports, vol. 11, pp. 1-10, 2021.
    [    CrossRef] [Google Scholar] [Publisher Link]
  42. Kuan Huang et al., “Shape-Adaptive Convolutional Operator for Breast Ultrasound Image Segmentation,” 2021 IEEE International Conference on Multimedia and Expo (ICME), Shenzhen, China, pp. 1-6, 2021.
    [    CrossRef] [Google Scholar] [Publisher Link]
  43. Ademola Enitan Ilesanmi, Utairat Chaumrattanakul, and Stanislav S. Makhanov, “A Method for Segmentation of Tumors in Breast Ultrasound Images using the Variant Enhanced Deep Learning,” Biocybernetics and Biomedical Engineering, vol. 41, no. 2, pp. 802-818, 2021.
    [    CrossRef] [Google Scholar] [Publisher Link]
  44. Wilfrido Gómez-Flores, and Wagner Coelho de Albuquerque Pereira, “A Comparative Study of Pre-trained Convolutional Neural Networks for Semantic Segmentation of Breast Tumors in Ultrasound,” Computers in Biology and Medicine, vol. 126, 2020.
    [    CrossRef] [Google Scholar] [Publisher Link]
  45. Sarita, and Rajeshwar Dass, “A Hybrid Model for Brain Tumor Classification via Deep Feature Extraction and Optimized Classifier Weighting,” Biomedical Physics & Engineering Express, vol. 11, no. 5, 2025.
    [CrossRef] [Google Scholar] [Publisher Link]
  46. Hailong Yang, Yinghao Liu, and Tian Xia, “Defect Detection Scheme of Pins for Aviation Connectors based on Image Segmentation and Improved RESNET-50,” International Journal of Image and Graphics, vol. 24, no. 1, 2024.
    [    CrossRef] [Google Scholar] [Publisher Link]
  47. Shoffan Saifullah et al., “Automatic Brain Tumor Segmentation: Advancing U-Net with ResNet50 Encoder for Precise Medical Image Analysis,” IEEE Access, vol. 13, pp. 43473-43489, 2025.
    [    CrossRef] [Google Scholar] [Publisher Link]
  48. Qinghua Huang et al., “Segmentation of Breast Ultrasound Image with Semantic Classification of Superpixels,” Medical Image Analysis, vol. 61, 2020.
    [    CrossRef] [Google Scholar] [Publisher Link]
  49. Ji Yang, Mehdi Faraji, and Anup Basu, “Robust Segmentation of Arterial Walls in Intravascular Ultrasound Images using Dual Path U-Net,” Ultrasonics, vol. 96, pp. 24-33, 2019.
    [    CrossRef] [Google Scholar] [Publisher Link]
  50. Animesh Karnewar et al., “ReLU Fields: The Little Non-Linearity that Could,” ACM SIGGRAPH 2022 Conference Proceedings, pp. 1-9, 2022.
    [    CrossRef] [Google Scholar] [Publisher Link]
  51. Jignesh Chowdary et al., “A Multi-Task Learning Framework for Automated Segmentation and Classification of Breast Tumors from Ultrasound Images,” Ultrasonic Imaging, vol. 44, no. 1, pp. 3-12, 2022.
    [    CrossRef] [Google Scholar] [Publisher Link]
  52. Niranjan Yadav, Rajeshwar Dass, and Jitendra Virmani, “Machine Learning-Based CAD System for Thyroid Tumour Characterisation using Ultrasound Images,” International Journal of Medical Engineering and Informatics, vol. 16, no. 6, pp. 547-559, 2024.
    [    CrossRef] [Google Scholar] [Publisher Link]
  53. Niranjan Yadav, Rajeshwar Dass, and Jitendra Virmani, “Assessment of Encoder-Decoder-Based Segmentation Models for Thyroid Ultrasound Images,” Medical & Biological Engineering & Computing, vol. 61, pp. 2159-2195, 2023.
    [    CrossRef] [Google Scholar] [Publisher Link]
  54. Sanggyu Kwak, “Are Only p-Values Less than 0.05 Significant? A p-Value Greater Than 0.05 Is Also Significant!,” Journal of Lipid and Atherosclerosis, vol. 12, no. 2, pp. 89-95, 2023.
    [    CrossRef] [Google Scholar] [Publisher Link]