Çoklu ölçekli morfoloji ve SVM tabanlı mamografi sınıflandırması

Abstract

Meme kanseri, kadınlarda en sık görülen ve erken tanı ile yaşam süresi ve kalitesinin önemli ölçüde artırılabildiği hastalıklardan biridir. Bu çalışmada, mamogram görüntülerinin normal - anormal olarak sınıflandırılması amacıyla etkili bir bilgisayar destekli teşhis (BDT) sistemi önerilmektedir. Tasarlanan sistem; çoklu ölçekli morfolojik top-hat dönüşümüne dayalı ön-işleme, dokusal ve uzaysal özellik çıkarımı, temel bileşenler analizi (TBA) ile boyut azaltımı ve destek vektör makineleri (DVM) tabanlı sınıflandırma aşamalarından oluşmaktadır. Ön işleme adımında görüntüyü iyileştirmek ve gürültüyü atmak için çoklu ölçekli morfolojik iyileştirme tekniği kullanılmıştır. Özellik çıkarımı aşamasında iyileştirilmiş mamogram görüntülerinden uzaysal özellikler ve dokusal özellikler çıkarılmıştır. Sınıflandırma aşamasında DVM kullanılmıştır. Deneysel çalışmalar MIAS mamografi veri tabanı üzerinde gerçekleştirilmiştir. Model performansı, daha güvenilir bir değerlendirme sağlamak için 5 katlı çapraz doğrulama yöntemi kullanılarak analiz edilmiştir. Sonuçlar, TBA uygulanmadan %91,86 doğruluk elde edilirken, TBA uygulandıktan sonra doğruluğun %93,02'ye yükseldiğini göstermiştir. Ayrıca, ROC analizi, boyut indirgeme sonrasında sınıflandırma performansında ve ayırt etme gücünde bir iyileşme olduğunu ortaya koymuştur (AUC = 0,8939). Sonuçlar, önerilen yöntemin özellikle sınırlı veri koşullarında rekabetçi, düşük maliyetli ve güvenilir bir performans sunduğunu ve meme kanserinin erken teşhisi için klinik karar destek sistemi olarak kullanılabileceğini göstermektedir.

Keywords

• Meme kanseri
• Momografi
• Bilgisayar Destekli Teşhis (BDT)
• Çok ölçekli top-hat dönüşümü
• Destek vektör makinesi (DVM)
• Temel Bileşenler Analizi (TBA)

Multiscale morphology and SVM-based mammogram classification

Abstract

Breast cancer is one of the most common diseases in women, and early diagnosis can significantly improve survival time and quality of life. This study proposes an effective computer-aided diagnosis (CAD) system for classifying mammogram images as normal or abnormal. The designed system consists of preprocessing based on multiscale morphological top-hat transformation, textural and spatial feature extraction, dimensionality reduction with principal component analysis (PCA), and support vector machine (SVM)-based classification stages. In the preprocessing step, a multiscale morphological enhancement technique was used to improve the image and remove noise. In the feature extraction stage, spatial and textural features were extracted from the enhanced mammogram images. In the classification stage, a SVM was used. Experimental studies were performed on the MIAS mammography database. Model performance was analyzed using a 5-fold cross-validation method to provide a more reliable assessment. The results showed that while an accuracy of 91.86% was obtained without PCA, the accuracy increased to 93.02% after PCA. Furthermore, ROC analysis revealed an improvement in classification performance and discrimination power after dimensionality reduction (AUC = 0.8939). The results demonstrate that the proposed method offers competitive, low-cost, and reliable performance, especially under limited data conditions, and can be used as a clinical decision support system for the early diagnosis of breast cancer.

Keywords

• Breast cancer
• Mammography
• Computer-aided diagnosis (CAD)
• Multiscale top-hat transform
• Support vector machine (SVM)
• Principal component analysis (PCA)

References

1. Zorukoş, S. (2008) Birinci Derece Yakınlarında Meme Kanseri Olan Kadınlar İçin Bilgi ve Destek Gereksinimleri Ölçeği’nin Türkçe’ye Uyarlanması, Yüksek Lisans Tezi, Dokuz Eylül University Institute of Health Science, İzmir, 82.
2. Ozmen, V. (2009) Meme hastalıklarının cerrahi tedavi kalitesi nasıl yükseltilebilir?, The Journal of Breast Health, 5(2): 119-121.
3. Giaquinto, A. N., Sung, H., Newman, L. A., Freedman, R. A., Smith, R. A., Star, J., Jemal, A., and Siegel, R. L. (2024) Breast cancer statistics 2024, CA: a cancer journal for clinicians, 74(6): 477-495.
4. Web Page: American Cancer Society, Breast Cancer Facts & Figures 2024–2025, https://www.cancer.org/research/cancer-facts-statistics/breast-cancer-facts-figures.html Last access date: 23.01.2026.
5. Web Page: Türkiye Kanser İstatistikleri, https://hsgm.saglik.gov.tr/depo/birimler/kanser-db/Dokumanlar/Istatistikler/Turkiye_Kanser_Istatistikleri_2020.pdf Last access date: 23.01.2026.
6. Nemade, V., Pathak, S., and Dubey, A.K. (2022) A systematic literature review of breast cancer diagnosis using machine intelligence techniques, Archives of Computational Methods in Engineering, 29(6): 4401-4430.
7. Shahid, M.S., and Imran, A. (2025) Breast cancer detection using deep learning techniques: challenges and future directions, Multimedia Tools and Applications, 84(6): 3257-3304.
8. Nass, S.J., Henderson, I.C., and Lashof, J.C. (2001) Mammography and Beyond: Developing Technologies for Early Detection of Breast Cancer, Washington: National Academy Press.

9. Bozek, J., Delac, K., and Grgic, M. (2008) Computer-Aided Detection of Breast Abnormalities in Digital Mammography, 50th International Symposium ELMAR, Zadar, Croatia.
10. Baert, A.L., Reiser, M.F., Hricak, H., and Knauth, M. (2010) Digital Mammography, Berlin: Springer-Verlag.
11. El-Naqa, I., Yang, Y., Wernick, M.N., Galatsanos, N.P., and Nishikawa, R.M. (2002) A Support Vector Machine Approach for Detection of Microcalcifications, IEEE Transactions on Medical Imaging, 21(12): 1552-1563.
12. Soltanian-Zadeh, H., Rafiee-Rad, F., and Pourabdollah-Nejad, D.S. (2004) Comparison of Multiewavelet, Wavelet, Haralick, and Shape Features for Microcalcification Classification in Mammograms, Pattern Recognition, 37(10): 1973-1986.
13. Fu, J.C., Lee, S.K., Wong, S.T.C., Yeh, J.Y., Wang, A.H., and Wu, H.K. (2005) Image Segmentation Feature Selection and Pattern Classification for Mammographic Microcalcifications, Computerized Medical Imaging and Graphics, 29(6): 419-429.
14. Zhang, X., Gao, X., and Wang, Y. (2009) MCs Detection with Combined Image Features and Twin Support Vector Machines, Journal of Computers, 4(3): 215-221.
15. Ragab, D.A., Sharkas, M., and Attallah, O. (2019) Breast Cancer Diagnosis Using an Efficient CAD System Based on Multiple Classifiers, Diagnostics, 9(4): 165.
16. Hekim, M., Aydın Yurdusev, A., and Oral, C. (2019) The Detection and Classification of Microcalcifications in the Visibility-Enhanced Mammograms Obtained by Using the Pixel Assignment-Based Spatial Filter, Advances in Electrical and Computer Engineering, 19(4): 73-82.
17. Fatima, N., Liu, L., Hong, S., and Ahmed, H. (2020) Prediction of Breast Cancer: Comparative Review of Machine Learning Techniques and Their Analysis, IEEE Access, 8: 150360-150376.
18. Ting, F., Tan, Y.J., and Sim, K.S. (2019) Convolutional Neural Network Improvement for Breast Cancer Classification, Expert Systems with Applications, 120: 103-115.
19. Hassan, S.A.A., Sayed, M.S., Abdalla, M.I., and Rashwan, M.A. (2020) Breast Cancer Masses Classification Using Deep Convolutional Neural Networks and Transfer Learning, Multimedia Tools and Applications, 79(41): 30735-30768.
20. El Houby, E.M., and Yassin, N.I. (2021) Malignant and Nonmalignant Classification of Breast Lesions in Mammograms Using Convolutional Neural Networks, Biomedical Signal Processing and Control, 70: 102954.
21. Laishram, R., and Rabidas, R. (2024) An Optimized Ensemble Classifier for Mammographic Mass Classification, Computers and Electrical Engineering, 119: 109488.
22. Suckling, J. (1994) The Mammographic Images Analysis Society Digital Mammogram Database, Excerpta Medica International Congress Series, York, England, 1069: 375-378.
23. Mukhopadhyay, S., and Chanda, B. (2000) A Multiscale Morphological Approach to Local Contrast Enhancement, Signal Processing, 80(4): 685-696.
24. Gonzalez, R.C., and Woods, R.E. (2002) Digital Image Processing, New Jersey: Prentice Hall.
25. Zheng, W., Yang, H., Sun, H. S., and Fan, H. Q. (2007) X-ray Image Enhancement Based on Multiscale Morphology, 1st International Conference on Bioinformatics and Biomedical Engineering, Wuhan, China.
26. Halkiotis, S., Botsis, T., and Rangoussi, M. (2007) Automatic Detection of Clustered Microcalcifications in Digital Mammograms Using Mathematical Morphology and Neural Networks, Signal Processing, 87(7): 1559-1568.
27. Haralick, R.M., Shanmugam, K., and Dinstein, I.H. (1973) Textural Features for Image Classification, IEEE Transactions on Systems, Man, and Cybernetics, 6: 610-621.
28. Cheng, H.D., and Cui, M. (2004) Mass Lesion Detection with a Fuzzy Neural Network, Pattern Recognition, 37(6): 1189-1200.
29. Zheng, B., Qian, W., and Clarke, L.P. (1996) Digital Mammography: Mixed Feature Neural Network with Spectral Entropy Decision for Detection of Microcalcifications, IEEE Transactions on Medical Imaging, 15(5): 589-597.
30. Oral, C., Aydın Yurdusev, A., and Bergil, E. (2019) Mamogramların Sınıflandırılmasında Dokusal Özelliklerin Etkileri, Dicle Üniversitesi Mühendislik Fakültesi Mühendislik Dergisi, 10(1): 23-33.
31. Liu, H., and Motoda, H. (2008) Computational Methods of Feature Selection, Boca Raton: Chapman & Hall/CRC press.
32. Tembey, M. (2003) Computer-Aided Diagnosis for Mammographic Microcalcification Clusters, Master’s Thesis, University of South Florida, USA, 98.
33. Meyer-Baese, A. (2004) Pattern Recognition for Medical Imaging, California: Elsevier Academic Press.
34. Costaridou, L. (2005) Medical Image Analysis Methods, USA: CRC Press.
35. Duda, R.O., Hart, P.E., and Stork, D.G. (2001) Pattern Classification, New York: Wiley-Interscience.
36. Smith, L.I. (2002) A Tutorial on Principal Component Analysis, https://www.iro.umontreal.ca/~pift6080/H08/documents/papers/pca_tutorial.pdf Last access date: 26.04.2011.
37. Jolliffe, I.T. (2002) Principal Component Analysis, Springer.
38. Vapnik, V.N. (1998) Statistical Learning Theory, Wiley-Interscience.
39. Zheng, N., and Xue, J. (2009) Statistical Learning and Pattern Analysis for Image and Video Processing, London: Springer.
40. Abe, S. (2005) Support Vector Machines for Pattern Classification, Springer.
41. Oral, C., and Sezgin, H. (2013) Effects of Dimension Reduction in Mammogram Classification, 8th International Conference on Electrical and Electronics Engineering (ELECO), Bursa, Türkiye, 630-633.