Mamografi Görüntülerinde Dışbükey Kitle Şekillerinin Hızlı Fourier Dönüşümünü (FFT) ve Zernike Momentlerini Kullanarak Kitle Algılama

Abstract

Bu çalışmada, mamografi görüntülerinde şüpheli bölge olarak tanımlanan bölgelerin dışbükey kitle sınırının Zernike momentlerinden ve Hızlı Fourier Dönüşümünden (FFT) faydalanarak kitle algılama uygulaması geliştirilir. Uygulamanın geliştirilmesi sırasında araştırmacıların kullanımına açık olan Mammografi Analiz Topluluğu (MIAS) veritabanı kullanılır. MIAS veritabanı, 322 adet 1024x1024 piksel çözünürlüklü normal, iyi huylu ve kötü huylu kanser mamografi görüntülerini içerir. Çalışmanın ilk aşamasında, görüntüler üzerinde gürültü azaltma ve görüntü iyileştirme işlemi yapılmaktadır. Şüpheli bölgelerle benzer özelliklere sahip olan pektoral kaslar görüntülerden ayrıştırılır. Ayrıştırma işleminden sonra, şüpheli bölgeleri netleştirmek için görüntüler kontrast yönünden iyileştirilir. Şüpheli bölgelerden, dışbükey kitle sınırının Zernike momentleri ve FFT'si hesaplanır ve her görüntü için öznitelik vektörleri elde edilir. Her bir görüntünün yeni öznitelik vektörü eğitim ve test kümelerine ayrılmış ve test kümesinin etiketleri %100 doğruluk ile elde edilmiştir.

Keywords

• Meme Kitlesi
• Pektoral Kas Ayrıştırma
• Destek Vektör Makinası
• Doğrusal Ayırma Analizi

Mass Detection Using the Zernike Moments and Fast Fourier Transform (FFT) of Convex Mass Shapes on Mammogram Images

Abstract

In this study, mass detection application is developed for mammograms from Zernike moments and Fast Fourier Transform (FFT) of convex mass boundary. During the development of the application, the Mammographic Image Analysis Society (MIAS) database, which is available to the researchers, is used. The MIAS database contains 322, 1024x1024 pixel resolution images of normal, benign, and malignant cancer. In the first phase of the study, noise reduction and image enhancement process is performed on the images. The pectoral muscles, which have similar features as region of interests (ROIs) are decomposed. After the decomposition process, images are enhanced by contrast to clarify ROIs. From ROIs, Zernike moments and FFT of convex mass boundary are calculated and feature vectors are obtained for each image. The new feature vector of each image was divided into training and test sets, and the labels of the test set were obtained with 100% accuracy.

Keywords

• Breast Mass
• Pectoral Muscle Removal
• Support Vector Machine
• Linear Discriminant Analysis

References

1. Jemal, A., Bray, F., Center, M. M., Ferlay, J., Ward, E., & Forman, D. (2011). Global cancer statistics. CA: a cancer journal for clinicians, 61(2), 69–90. https://doi.org/10.3322/caac.20107
2. Zhang, Z., Lu, J., & Yip J. (2008). Computer aided mammography. In: Proceedings of Computing and Engineering Annual Researchers' Conference 2008: CEARC’08, University of Huddersfield, Huddersfield, 125-130. ISBN 978-1-86218-067-3.
3. Divyashree, B. & Kumar, G. (2021). Breast Cancer Mass Detection in Mammograms Using Gray Difference Weight and MSER Detector. SN Computer Science, 2. 10.1007/s42979-021-00452-8.
4. Lbachir, I. A., Daoudi, I., & Tallal, S. (2020). Automatic computer-aided diagnosis system for mass detection and classification in mammography. Multimedia Tools and Applications, 80(6), 9493–9525. https://doi.org/10.1007/s11042-020-09991-3
5. Sarangi, S., Rath, N. P., & Sahoo, H. K. (2021). Mammogram mass segmentation and detection using Legendre neural network-based optimal threshold. Medical & Biological Engineering & Computing, 59(4), 947–955. https://doi.org/10.1007/s11517-021-02348-4
6. Braz Junior, G., da Rocha, S. V., de Almeida, J. D. S., de Paiva, A. C., Silva, A. C., & Gattass, M. (2018). Breast cancer detection in mammography using spatial diversity, geostatistics, and concave geometry. Multimedia Tools and Applications, 78(10), 13005–13031. https://doi.org/10.1007/s11042-018-6259-z
7. Dhungel, N., Carneiro, G., & Bradley, A. P. (2017). A deep learning approach for the analysis of masses in mammograms with minimal user intervention. Medical Image Analysis, 37, 114–128. https://doi.org/10.1016/j.media.2017.01.009
8. Zhu, W., Lou, Q., Vang, Y. S., & Xie, X. (2017). Deep Multi-instance Networks with Sparse Label Assignment for Whole Mammogram Classification. Medical Image Computing and Computer Assisted Intervention − MICCAI 2017, 603–611. https://doi.org/10.1007/978-3-319-66179-7_69

9. Platania, R., Shams, S., Yang, S., Zhang, J., Lee, K., & Park, S. J. (2017). Automated Breast Cancer Diagnosis Using Deep Learning and Region of Interest Detection (BC-DROID). Proceedings of the 8th ACM International Conference on Bioinformatics, Computational Biology, and Health Informatics, 536–543. https://doi.org/10.1145/3107411.3107484
10. Suckling, J., et al (1994). The Mammographic Image Analysis Society Digital Mammogram Database Exerpta Medica. International Congress Series, 1069, 375-378.
11. Esener, İ., Ergi̇n, S., & Yüksel, T. (2018). A novel multistage system for the detection and removal of pectoral muscles in mammograms. Turkish Journal of Electrical Engineering and Computer Science, 26 (1), 35-49.
12. Gallagher, N., & Wise, G. (1981). A theoretical analysis of the properties of median filters. IEEE Transactions on Acoustics, Speech, and Signal Processing, 29(6), 1136–1141. https://doi.org/10.1109/tassp.1981.1163708
13. Nagi, J., Kareem, S. A., Nagi, F., & Ahmed, S. K. (2010). Automated breast profile segmentation for ROI detection using digital mammograms. 2010 IEEE EMBS Conference on Biomedical Engineering and Sciences (IECBES), 87-92. doi:10.1109/iecbes.2010.5742205.
14. Khotanzad, A., & Hong, Y. (1990). Invariant image recognition by Zernike moments. IEEE Transactions on Pattern Analysis and Machine Intelligence, 12(5), 489-497. doi:10.1109/34.55109