Mamografi görüntülerindeki anormalliklerin yerel ikili örüntü ve varyantları kullanılarak sınıflandırılması

Meme kanseri teşhisinde kullanılan mamografilerdeki anormalliklerin sınıflandırılması için makine öğrenmearaştırmaları büyük önem arz etmektedir. Bu çalışmada Curated Breast Imaging Subset of Digital Database forScreening Mammography (CBIS-DDSM) görüntü tabanındaki kitleli ve kalsifikasyonlu mamografi görüntülerisınıflandırılmıştır. Veri setindeki görüntülerden Yerel İkili Örüntü(YİÖ), Yerel Türev Örüntü, Yerel DörtlüÖrüntü(YDÖ), ve Gürültüye Dirençli Yerel İkili Örüntü yöntemleri ile doku öznitelikleri çıkarılmıştır. Öznitelikçıkarım yöntemlerinden yerel çarpıklık örüntü tabanlı ayrıntılı histogram yöntemiyle de öznitelik çıkarımıyapılmıştır. Daha sonra öznitelik vektörleri doğrusal ve radyal tabanlı fonksiyon kernel destek vektörmakineleri(DVM) ve yapay sinir ağları (YSA) kullanılarak sınıflandırılmıştır. Eğitim ve doğrulama verisi için 5-kez çapraz doğrulama yöntemi uygulanmıştır. En yüksek sınıflandırma performansı veren eşik seviyeleri vepencere boyutları her bir öznitelik çıkarım yöntemi için belirlenmiştir. Öznitelik çıkarımı için gerekli olan sürelertablo halinde verilmiştir. Öznitelik çıkarım yöntemi olarak farklı çap ve nokta sayısı ile hesaplanmış YİÖ vektörlerifüzyonu ve sınıflandırıcı olarak 2 gizli katmanlı YSA kullanılması durumunda test verisi için %85.74 başarı oranıelde edilmiştir. Elde edilen başarı oranları literatürdeki makine öğrenmesi sonuçlarına göre yüksek ve derinöğrenme sonuçları ile kıyaslanabilir sonuçlardır.

Classification of abnormality in breast cancer mammography using local binary patterns and its variants

Machine learning research for the classification of breast cancer abnormalities is very crucial. In this study, mass and calcification abnormalities of mammograms in the Curated Breast Imaging Subset of Digital Database for Screening Mammography (CBIS-DDSM) were classified. Local binary pattern (LBP), local derivative pattern, local tetra pattern (LTrP), and noise resistant local binary pattern features were extracted from the images in the data set. Histogram refinement methods of LBP and its variants based on local skewness pattern were also extracted. Feature vectors were then classified by using linear and radial basis function kernel support vector machines (SVM) and an artificial neural network (ANN). 5-fold cross-validation was conducted. The highest performance classification conditions including threshold level and the window size were reported for LBP, its variants, and the histogram refinement methods. Feature extraction time requirements for LBP and its variants were reported. Test accuracy of 85.74 was achieved when feature vectors were extracted at different radius and number of points and fused and classified by using 2 hidden-layer neural network. The test accuracies obtained in this study are better than the machine learning approaches in the literature and comparable with the deep learning results.

PDF

___

[1] Özmen V. 2014. Breast Cancer in Turkey: Clinical and Histopathological Characteristics (Analysis of 13.240 Patients). Journal of Breast Health, 10 (2): 98-105.
[2] Xi P., Shu C., Goubran R. 2018. Abnormality Detection in Mammography using Deep Convolutional Neural Networks. 2018 IEEE International Symposium on Medical Measurements and Applications, 11-13 Haziran, Roma, İtalya, 1-6.
[3] Tang J., Rangayyan R.M., Xu J., Naqa I.E., Yang Y. 2009. Computer aided detection and diagnosis of breast cancer with mammography: Recent advances. IEEE Transactions on Information Technology in Biomedicine, 13 (2): 236-251.
[4] Miller R.G. 2001. Breast cancer screening. Journal of General Internal Medicine, 16 (3): 206-207.
[5] Lee R.S., Gimenez F., Hoogi A., Rubin D. 2016. The Cancer Imaging Archive. "Curated Breast Imaging Subset of DDSM". http://dx.doi.org/10.7937/K9/TCIA.2016.7O02S9CY (Erişim Tarihi: 01.02.2019).
[5] Lee R.S., Gimenez F., Hoogi A., Miyake K.K., Gorovoy M., Rubin D.L. 2017. A curated mammography data set for use in computer-aided detection and diagnosis research. Scientific Data, 4 (170177): 1-9.
[6] Clark K., Vendt B., Smith K., Freymann J., Kirby J., Koppel P., Moore S., Phillips S., Maffitt D., Pringle M., Tarbox L., Prior F. 2013. The Cancer Imaging Archive (TCIA): Maintaining and Operating a Public Information Repository. Journal of Digital Imaging, 26 (6): 1045-1057.
[7] Suckling J., Parker J., Dance D., Astley S., Hutt I., Boggis C., Savage J. 1994. The Mammographic Image Analysis Society digital mammogram database. 2nd International Workshop on Digital Mammography, Amsterdam, 375-378.
[8] Ackerman L.V., Gose E.E. 1972. Breast lesion classification by computer and xeroradiograph. Cancer, 30 (4): 1025-1035.
[9] Kutluk S., Günsel B. 2013. Tissue density classification in mammographic images using local features. 21st Signal Processing and Communications Applications Conference, 24-26 Nisan, Haspolat, Türkiye, 1-4.
[10] Talha M. 2016. Classification of mammograms for breast cancer detection using fusion of discrete cosine transform and discrete wavelet transform features. Biomedical Research, 27 (2): 322-327.
[11] Spanhol F.A., Oliveira L.S., Cavalin P.R., Petitjean C., Heutte L. 2017. Deep features for breast cancer histopathological image classification. International Conference on Systems, Man, and Cybernetics, 5-8 Ekim, Alberta, Kanada, 1868-1873.
[12] Qayyum A., Basit A. 2016. Automatic breast segmentation and cancer detection via SVM in mammograms. International Conference on Emerging Technologies. 18-19 Ekim, İslamabad, Pakistan, 1-6.
[13] Ojala T., Pietikäinen M., Mäenpää T. 2002. Multiresolution gray-scale and rotation invariant texture classification with local binary patterns. IEEE Transactions on Pattern Analysis and Machine Intelligence, 24 (7): 971-987.
[14] Burçin K., Vasif N.V. 2011. Down syndrome recognition using local binary patterns and statistical evaluation of the system. Expert Systems with Applications, 38 (7): 8690-8695.
[15] Kaya Y., Uyar M., Tekin R., Yıldırım S. 2014. 1D-local binary pattern based feature extraction for classification of epileptic EEG signals. Applied Mathematics and Computation, 243 (2014): 209-219.
[16] Oppedal K., Eftestøl T., Engan K., Beyer M.K., Aarsland D. 2015. Classifying dementia using local binary patterns from different regions in magnetic resonance images. International Journal of Biomedical Imaging, 572567: 1-14.
[17] Lenc L, Kral P. 2016. LBP Features for Breast Cancer Detection. IEEE International Conference on Image Processing, 25-28 Eylül, Phoenix, Arizona, ABD, 2643-2647.
[18] Nahid A.A., Kong Y. 2018. Histopathological Breast-Image Classification Using Local and Frequency Domains by Convolutional Neural Network. Information, 9 (1): 19-45.
[19] Esener İ.I., Ergin S., Yüksel T. 2015. Göğüs Kanseri Teşhisinde Bir Öznitelik Seçim Analizi. Tıp Teknolojileri Ulusal Kongresi, 15-18 Ekim, Muğla, 416-419.
[20] Kadiroğlu Z., Şengür A., Deniz E. 2018. Classifıcation of Histopathological Breast Cancer Images With Low Level Texture Features. International Engineering and Natural Sciences Conference, Kasım 2018, Diyarbakır, 1-8.
[21] Reston V.A. 2003. ACR BI-RADS-mammography, ultrasound and magnetic resonance imaging. 4th ed., American College of Radiology.
[22] Ojala T., Pietikäinen M., Mäenpää T. 2000. Gray Scale and Rotation Invariant Texture Classification with Local Binary Patterns. European Conference on Computer Vision, Berlin, 404-420.
[23] Zhang B., Gao Y., Zhao S., Liu J. 2010. Local derivative pattern versus local binary pattern: face recognition with high-order local pattern descriptor. IEEE Transactions on Image Processing, 19 (2): 533-544.
[24] Murala S., Maheshwari R.P., Balasubramanian R. 2012. Local tetra patterns: a new feature descriptor for content-based image retrieval. IEEE Transactions on Image Processing, 21 (5): 2874-2886.
[25] Ren J., Jiang X., Yuan J. 2013. Noise-resistant local binary pattern with an embedded errorcorrection mechanism. IEEE Transactions on Image Processing, 22 (10): 4049-4060.
[26] Tiwari A.T., Kanhangad V., Pachori R.B. 2017. Histogram refinement for texture descriptor based image retrieval. Signal Processing: Image Communication, 53: 73-85.
[27] Cristianini N, Shawe-Taylor J. 2000. An Introduction to Support Vector Machines and other kernel based learning methods. AI Magazine, Cambridge University Press, Cambridge.
[28] Schölkopf B., Smola, A.J. 2001. Learning with kernels: Support Vector Machines, Regularization, Optimization and Beyond. The MIT Press, 644s.
[29] Suykens J.A.K., Van Gestel T., De Brabanter J., De Moor B., Vandewalle J. 2002. Least Squares Support Vector Machines. World Scientific, Singapur, 291s.
[30] Marquardt D. 1963. An Algorithm for Least-Squares Estimation of Nonlinear Parameters. SIAM Journal on Applied Mathematics, 11 (2): 431-441.
[31] Hagan M.T., Menhaj M. 1994. Training feed-forward networks with the Marquardt algorithm. IEEE Transactions on Neural Networks, 5 (6): 989–993.
[32] Tsochatzidis L., Costaridou L., Pratikakis I. 2019. Deep Learning for Breast Cancer Diagnosis from Mammograms-A Comparative Study. Journal of Imaging, 5 (37): 1-11.
[33] Alhakeem Z., Jang S.A. 2019. Convolution-Free Lbp-Hog Descriptor for Mammogram Classification. arXiv preprint arXiv: 1904.00187: 1-5.