TEXTURAL FEATURE BASED REGION OF INTEREST DETECTION USING ANN FROM MAMMOGRAMS

Year 2020, Volume: 8 Issue: 5, 133 - 141, 29.12.2020

Sena Büşra Yengeç Taşdemir , Kasım Taşdemir , Zafer Aydın

https://doi.org/10.21923/jesd.827131

Cited By: 1

Abstract

Radiologists’ Type I error rate of Breast Cancer Detection from mammography images can reach up to thirty percent. In this study to assist radiology experts, a new Computer Aided Detection (CAD) system is proposed in order to increase the detection rate of Breast Cancer. A CAD system distinguishes the cancerous regions from normal tissues. In the proposed system, Haralick and HOG features are extracted from two-dimensional Wavelet transformed images which are enhanced by the CLAHE method. PCA algorithm is employed to select the extracted features. The selected features are given as input to a multi-layer perceptron (MLP) architecture. A detection accuracy of 81% is achieved when Adam optimization is used. In addition, various machine learning and deep learning methods have been implemented for comparison. When limited number of samples are used, the detection success of deep learning methods decreases even if transfer learning is employed. On the contrary, conventional computer vision methods give more successful results when appropriate combination of preprocessing, feature selection and machine learning algorithms are selected.

Keywords

Computer Aided Cancer Detection, Machine Learning, Computer Vision, Transfer Learning with Deep Learning

YSA KULLANILARAK MAMOGRAMLARDAN DOKUSAL ÖZNİTELİK TABANLI MEME KANSERİ İLGİ BÖLGESİ TESPİTİ

Abstract

Radyoloji uzmanlarının mamografi görüntülerine bakarak yaptığı meme kanseri teşhislerinde tip bir hata oranı yüzde otuzlara kadar çıkmaktadır. Kanserin teşhis başarısını artırmak adına bu çalışmada uzmanlara yardımcı olacak yeni bir Bilgisayar Yardımlı Teşhis sistemi, kanserli ve normal dokuyu ayırt etmek için önerilmektedir. Önerilen sistemde kontrast limitli histogram eşitleme (CLAHE) yöntemiyle iyileştirilen görüntülerin iki boyutlu parçacık dönüşümlerinden (2B–DWT) Haralick ve HOG öznitelikleri çıkarılmıştır. Özniteliklerin sayısını azaltması için temel bileşenler analizi (PCA) algoritması kullanılmıştır. Seçilen öznitelikler çok katmanlı algılayıcı (MLP) mimari yapısına sahip yapay sinir ağına (YSA) girdi olarak verilmiştir. Çok katmanlı algılayıcı üzerinde Adam eniyileme yapıldığında %81 tespit doğruluğu yakalanmıştır. Ayrıca, diğer bir çok temel makine öğrenmesi ve derin öğrenme yöntemleri denenerek karşılaştırma sonuçları detaylı olarak sunulmuştur. Sınırlı sayıda veri kümesi kullanıldığında transfer öğrenim kullanılsa dahi derin öğrenme yöntemlerinin tespit başarısı azalmıştır. Buna karşılık doğru ön işleme, öznitelik seçilimi ve makine öğrenmesi yaklaşımları kullanıldığı zaman geleneksel bilgisayarlı görü yöntemleri daha başarılı sonuçlar vermiştir

Keywords

Bilgisayar Yardımlı Kanser Tespiti, Makine Öğrenmesi, Bigisayarlı Görü, Transfer Öğrenimi ile Derin Öğrenme

References

• “The mini-MIAS database of mammograms.” [Online]. Available: http://peipa.essex.ac.uk/info/mias.html. [Accessed: 10-Feb-2019].
• Abadi, M., Barham, P., Chen, J., Chen, Z., Davis, A., Dean, J., ... & Kudlur, M. (2016). Tensorflow: A system for large-scale machine learning. In 12th {USENIX} symposium on operating systems design and implementation ({OSDI} 16) (pp. 265-283).
• Becker, A. S., Marcon, M., Ghafoor, S., Wurnig, M. C., Frauenfelder, T., & Boss, A. (2017). Deep learning in mammography: diagnostic accuracy of a multipurpose image analysis software in the detection of breast cancer. Investigative radiology, 52(7), 434-440.
• Coelho, L. P. (2012). Mahotas: Open source software for scriptable computer vision. arXiv preprint arXiv:1211.4907.
• Gedik, N. (2016). A new feature extraction method based on multi-resolution representations of mammograms. Applied Soft Computing, 44, 128-133.
• G. R. Lee et al., “PyWavelets/pywt: PyWavelets v1.0.0,” Aug. 2018.
• Hall, M., Frank, E., Holmes, G., Pfahringer, B., Reutemann, P., & Witten, I. H. (2009). The WEKA data mining software: an update. ACM SIGKDD explorations newsletter, 11(1), 10-18.
• Jiao, Z., Gao, X., Wang, Y., & Li, J. (2018). A parasitic metric learning net for breast mass classification based on mammography. Pattern Recognition, 75, 292-301.
• Oeffinger, K. C., Fontham, E. T., Etzioni, R., Herzig, A., Michaelson, J. S., Shih, Y. C. T., ... & Wolf, A. M. (2015). Breast cancer screening for women at average risk: 2015 guideline update from the American Cancer Society. Jama, 314(15), 1599-1614.
• Pedro, R. W. D., Machado-Lima, A., & Nunes, F. L. (2019). Is mass classification in mammograms a solved problem?-a critical review over the last 20 years. Expert Systems with Applications, 119, 90-103.
• Raghu, M., Zhang, C., Kleinberg, J., & Bengio, S. (2019). Transfusion: Understanding transfer learning for medical imaging. In Advances in neural information processing systems (pp. 3347-3357).
• Rodriguez-Ruiz, A., Mordang, J. J., Karssemeijer, N., Sechopoulos, I., & Mann, R. M. (2018, July). Can radiologists improve their breast cancer detection in mammography when using a deep learning based computer system as decision support?. In 14th International Workshop on Breast Imaging (IWBI 2018) (Vol. 10718, p. 1071803). International Society for Optics and Photonics.
• Saki, F., Tahmasbi, A., Soltanian-Zadeh, H., & Shokouhi, S. B. (2013). Fast opposite weight learning rules with application in breast cancer diagnosis. Computers in biology and medicine, 43(1), 32-41.
• Šerifović-Trbalić, A., Trbalić, A., Demirović, D., Prljača, N., & Cattin, P. C. (2014, May). Classification of benign and malignant masses in breast mammograms. In 2014 37th International Convention on Information and Communication Technology, Electronics and Microelectronics (MIPRO) (pp. 228-233). IEEE.
• Tahmasbi, A., Saki, F. and Shokouhi, S. B. (2011) ''Classification of benign and malignant masses based on Zernike moments'', Computers in Biology and Medicine, 41, 726–735.
• Tasdemir, S. B. Y., Tasdemir, K., & Aydin, Z. (2018, December). ROI Detection in Mammogram Images using Wavelet-Based Haralick and HOG Features. In 2018 17th IEEE International Conference on Machine Learning and Applications (ICMLA) (pp. 105-109). IEEE.
• Yassin, N. I., Omran, S., El Houby, E. M., & Allam, H. (2018). Machine learning techniques for breast cancer computer aided diagnosis using different image modalities: A systematic review. Computer methods and programs in biomedicine, 156, 25-45.
• Yengec-Tasdemir, S. B., Tasdemir, K., & Aydin, Z. (2020). A review of mammographic region of interest classification. Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, e1357.