A multi-scale structural-texture framework for mammographic image analysis

Authors

https://doi.org/10.22105/thi.v3i2.53

Abstract

Accurate analysis of mammographic images requires effective representation of tissue characteristics while maintaining robustness when the available dataset is limited. This study presents a multi-scale structural and texture analysis framework for mammographic image analysis using a Support Vector Machine (SVM). Manually extracted Regions of Interest (ROIs) were characterized using Fractal Dimension (FD), Lacunarity (LAC), and multi-scale Gray-Level Co-Occurrence Matrix (GLCM) features. Texture information was extracted at three spatial scales, namely 3×3, 5×5, and 7×7, to capture local and broader spatial characteristics. The feature representation obtained is made up of 14 features, which were used as input to SVM classifier kinds of algorithms. The dataset comprises a total of 100 samples, i.e., 50 benign and 50 malignant samples. The dataset was split randomly into the training/testing dataset with the training data representing 70% of the total samples and testing data representing 30% of the total dataset. To assess the model more reliably, five-fold cross-validation was conducted within the training set. This entire process was repeated 10 times using different random partitions of the data. Across these 10 repetitions, the proposed framework achieved a mean classification accuracy of 97.33%. The findings demonstrate that combining structural and multi-scale texture descriptors within a compact 14-dimensional feature space can provide effective mammographic image analysis without relying on the high architectural complexity and extensive parameterization of deep-learning models. Nevertheless, validation using larger and independent datasets is required to further assess the robustness and generalizability of the proposed framework.

Keywords:

Breast cancer, Fractal dimension, Lacunarity, Multi-scale gray-level co-occurrence matrix, Support vector machine

References

  1. [1] Ebrahimi, A., Arian, A., Akbari, S. A. L. I., Ahmadinejad, N., & Ahmadinejad, N. (2022). Diagnostic accuracy of opportunistic breast cancer screening based on mammography in Iran. Iranian Journal of Radiology, 19(3), 1–7. https://www.sid.ir/paper/1145624/
  2. [2] Diaz, O., Rodriguez-Ruiz, A., & Sechopoulos, I. (2024). Artificial intelligence for breast cancer detection: Technology, challenges, and prospects. European Journal of Radiology, 175, 111457. https://doi.org/10.1016/j.ejrad.2024.111457
  3. [3] Tai, S. C., Chen, Z. S., & Tsai, W. T. (2013). An automatic mass detection system in mammograms based on complex texture features. IEEE Journal of Biomedical and Health Informatics, 18(2), 618–627. https://doi.org/10.1109/JBHI.2013.2279097
  4. [4] Ergin, S., Esener, İ. I., & Yüksel, T. (2016). A genuine GLCM-based feature extraction for breast tissue classification on mammograms. International Journal of Intelligent Systems and Applications in Engineering, 4(Special Issue-1), 124-129. https://doi.org/10.18201/ijisae.269453
  5. [5] Letchumanan, N., Wong, J. H. D., Tan, L. K., Ab Mumin, N., Ng, W. L., Chan, W. Y., & Rahmat, K. (2023). A radiomics study: Classification of breast lesions by textural features from mammography images. Journal of Digital Imaging, 36(4), 1533–1540. https://doi.org/10.1007/s10278-022-00753-1
  6. [6] Qi, Y J., Su, G. H., You, C., Zhang, X., Xiao, Y., Jiang, Y. Z., & Shao, Z. M. (2024). Radiomics in breast cancer: Current advances and future directions. Cell Reports Medicine, 5(9), 1–25. https://doi.org/10.1016/j.xcrm.2024.101719.
  7. [7] Zhang, X., Zhang, S., Bu, Z., Ma, L., & Huang, J. (2023). Texture feature dimensionality reduction-based mammography classification using random forest. Journal of Computational Methods in Sciences and Engineering, 23(3), 1537–1545. https://doi.org/10.3233/JCM-226669
  8. [8] Guo, J., Yuan, H., Shi, B., Zheng, X., Zhang, Z., Li, H., & Sato, Y. (2024). A novel breast cancer image classification model based on multiscale texture feature analysis and dynamic learning. Scientific Reports, 14(1), 7216. https://doi.org/10.1038/s41598-024-57891-5
  9. [9] Guo, Q., Shao, J., & Ruiz, V. F. (2009). Characterization and classification of tumor lesions using computerized fractal-based texture analysis and support vector machines in digital mammograms. International Journal of Computer Assisted Radiology and Surgery, 4(1), 11–25. https://doi.org/10.1007/s11548-008-0276-8
  10. [10] Ancy, C. A., & Nair, L. S. (2017). An efficient CAD for detection of tumour in mammograms using SVM. 2017 international conference on communication and signal processing (ICCSP) (pp. 1431–1435). IEEE. https://doi.org/10.1109/ICCSP.2017.8286621
  11. [11] Dong, M., Lu, X., Ma, Y., Guo, Y., Ma, Y., & Wang, K. (2015). An efficient approach for automated mass segmentation and classification in mammograms. Journal of Digital Imaging, 28(5), 613–625. https://doi.org/10.1007/s10278-015-9778-4
  12. [12] Deshmukh, Y. S., Kumar, P., Karan, R., & Singh, S. K. (2021). Retracted: Breast cancer detection-based feature optimization using firefly algorithm and ensemble classifier. 2021 international conference on artificial intelligence and smart systems (ICAIS) (pp. 1048–1054). IEEE. https://doi.org/10.1109/ICAIS50930.2021.9395788
  13. [13] Elahi, R., & Nazari, M. (2024). An updated overview of radiomics-based artificial intelligence (AI) methods in breast cancer screening and diagnosis. Radiological Physics and Technology, 17(4), 795–818. https://doi.org/10.1007/s12194-024-00842-6
  14. [14] Alexey, K., Kolchev, A. A., Dmitry, P., Pasynkov, D. V, Ivan, E., Egoshin, I. A., … & Olga, P. (2024). Classification of benign and malignant solid breast lesions on the ultrasound images based on the textural features: The importance of the perifocal lesion area. Computer Optics, 48(1), 157–165. https://doi.org/10.18287/2412-6179-CO-1244
  15. [15] Suckling, J. (1994). The mammographic images analysis society digital mammogram database. Digital mammography: proceedings of the 2nd international workshop on digital mammography (pp. 375–378). Excerpta Medica. https://www.repository.cam.ac.uk/items/b6a97f0c-3b9b-40ad-8f18-3d121eef1459?utm_source=chatgpt.com
  16. [16] Bisoi, A. K., & Mishra, J. (2001). On calculation of fractal dimension of images. Pattern Recognition Letters, 22(6–7), 631–637. https://doi.org/10.1016/S0167-8655(00)00132-X
  17. [17] Pezeshki, H., Rastgarpour, M., Sharifi, A., & Yazdani, S. (2019). Extraction of spiculated parts of mammogram tumors to improve accuracy of classification. Multimedia Tools and Applications, 78(14), 19979–20003. https://doi.org/10.1007/s11042-019-7185-4
  18. [18] Mohanty, F., Rup, S., Dash, B., Majhi, B., & Swamy, M. N. S. (2019). Mammogram classification using contourlet features with forest optimization-based feature selection approach. Multimedia Tools and Applications, 78(10), 12805–12834. https://doi.org/10.1007/s11042-018-5804-0
  19. [19] Xie, W., Li, Y., & Ma, Y. (2016). Breast mass classification in digital mammography based on extreme learning machine. Neurocomputing, 173, 930–941. https://doi.org/10.1016/j.neucom.2015.08.048
  20. [20] Pashoutan, S., Shokouhi, S. B., & Pashoutan, M. (2017). Automatic breast tumor classification using a level set method and feature extraction in mammography. 2017 24th national and 2nd international Iranian conference on biomedical engineering (ICBME) (pp. 1–6). IEEE. https://doi.org/10.1109/ICBME.2017.8430285
  21. [21] Tatikonda, K. C., Bhuma, C. M., & Samayamantula, S. K. (2018). The analysis of digital mammograms using HOG and GLCM features. 2018 9th international conference on computing, communication and networking technologies (ICCCNT) (pp. 1–7). IEEE. https://doi.org/10.1109/ICCCNT.2018.8493809
  22. [22] Muduli, D., Dash, R., & Majhi, B. (2022). Automated diagnosis of breast cancer using multi-modal datasets: A deep convolution neural network based approach. Biomedical Signal Processing and Control, 71, 102825. https://doi.org/10.1016/j.bspc.2021.102825

Published

2026-06-13

Issue

Vol. 3 No. 2 (2026): Trends in Health Informatics

Section

Articles

License

Copyright (c) 2026 Trends in Health Informatics

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

How to Cite

Mar Cornelio, O. . (2026). A multi-scale structural-texture framework for mammographic image analysis. Trends in Health Informatics, 3(2), 113-123. https://doi.org/10.22105/thi.v3i2.53

Similar Articles

1-10 of 21

You may also start an advanced similarity search for this article.