Meme Kanseri Tespiti için Destek Vektör Makinası ile Alexnet Kullanarak Transfer Öğrenimi

Year 2020, Ejosat Special Issue 2020 (ICCEES), 423 - 430, 05.10.2020

Sema Abdulghani , Ahmed Fadhil Seyfettin Sinan Gültekin

https://doi.org/10.31590/ejosat.806679

Abstract

Meme kanseri, şu anda dünya çapında kadın ölümlerinin önde gelen nedenlerinden biridir. Meme kanseri teşhisi için bilgisayar destekli teşhis sistemleri geliştirmek, son yıllarda birçok araştırmacı için ilgi çekici bir sorun haline geldi. Araştırmacılar, büyük bir başarı elde eden Evrişimli Sinir Ağları (CNN'ler) dâhil olmak üzere sınıflandırma problemleri için derin Öğrenme tekniklerine odaklandılar. CNN'ler, özellikle biyomedikal görüntü işleme görevlerinde deneysel başarılar elde eden, araştırma topluluğu ve endüstriden dikkat çeken özel bir derin, ileri beslemeli ağ türüdür. Bu çalışmada, meme kanseri histopatolojik görüntülerini kamuya açık (BreakHis veri seti) sınıflandırmak için önceden eğitilmiş bir CNN modelini uyarlayan transfer öğrenme ve derin özellik çıkarma yöntemleri kullanılmıştır. AlexNet modeli bu çalışmada yama stratejisi ile ele alındı ve daha fazla ince ayar için önceden eğitilmiş AlexNet kullanıldı. Elde edilen özellikler daha sonra destek vektör makineleri (SVM) kullanılarak sınıflandırıldı. Değerlendirme sonuçları, SVM sınıflandırıcısı ile önceden eğitilmiş Alexnet'in, farklı büyütme faktörleri için beş kat çapraz doğrulama tekniği kullanarak 92 % ile 96 % arasında bir doğruluk sağladığını göstermektedir.

Keywords

Meme Kanseri, Evreşimli sinir ağları, Transfer Öğrenimi, Destek Vektör Makinesi

Transfer Learning using Alexnet with Support Vector Machine for Breast Cancer Detection

Abstract

Breast cancer is one of the leading causes of women death worldwide currently. Developing a computer-aided diagnosis system for breast cancer detection became an interesting problem for many researchers in recent years. Researchers focused on deep learning techniques for classification problems, including Convolutional Neural Networks (CNNs), which achieved great success. CNN is a particular type of deep, feedforward network that has gained attention from the research community and achieved great successes, especially in biomedical image processing. In this paper, transfer learning and deep feature extraction methods are used which adapt a pre-trained CNN model to classify breast cancer histopathological images from the publically available (BreakHis dataset). The data set includes both benign and malignant images with four different magnification factors. A patch strategy method proposed based on the extraction of image patches for training the CNN and the combination of these patches for final classification. AlexNet model is considered in this work with patch strategy, and pre-trained AlexNet is used for further fine-tuning. The obtained features are then classified by using support vector machines (SVM). The evaluation results show that the pre-trained Alexnet with SVM classification and patch strategy yields the best accuracy. Accuracy between 92% and 96% was achieved using five-fold cross-validation technique for different magnification factors.

Keywords

Breast Cancer, Convolutional Neural Network, Alexnet, Transfer Learning, Support Vector Machine

References

• Abd Almisreb, A., et al. (2018). Utilizing AlexNet deep transfer learning for ear recognition. 2018. Fourth International Conference on Information Retrieval and Knowledge Management (CAMP), IEEE.
• Berrar, D. J. E. o. B. and C. Biology (2019). "Cross-validation." 1: 542-545.
• Bottou, L. (2012). Stochastic gradient descent tricks. Neural networks: Tricks of the trade, Springer: 421-436.
• Boyle, P. and B. Levin (2008). World cancer report 2008, IARC Press, International Agency for Research on Cancer.
• Deniz, E., et al. (2018). "Transfer learning based histopathologic image classification for breast cancer detection." 6(1): 18.
• Duda, R. O., et al. (2012). Pattern classification, John Wiley & Sons.
• Fadhil, A. F. (2014). Formulation of detection strategies in images, Southern Illinois University at Carbondale.
• Hafemann, L. G., et al. (2014). Forest species recognition using deep convolutional neural networks. 2014 22nd International Conference on Pattern Recognition, IEEE.
• Huang, Z., et al. (2017). "Transfer learning with deep convolutional neural network for SAR target classification with limited labeled data." 9(9): 907.
• Kassani, S. H., et al. (2019). "Breast cancer diagnosis with transfer learning and global pooling."
• Krizhevsky, A., et al. (2012). Imagenet classification with deep convolutional neural networks. Advances in neural information processing systems.
• Lakhani, S., et al. (2012). "WHO Classification of Tumours of the Breast, ed 4. Lyon."
• Niu, X.-X. and C. Y. J. P. R. Suen (2012). "A novel hybrid CNN–SVM classifier for recognizing handwritten digits." 45(4): 1318-1325.
• Spanhol, F. A., et al. (2017). Deep features for breast cancer histopathological image classification. 2017 IEEE International Conference on Systems, Man, and Cybernetics (SMC), IEEE.
• Spanhol, F. A., et al. (2016). Breast cancer histopathological image classification using convolutional neural networks. 2016 international joint conference on neural networks (IJCNN), IEEE.
• Spanhol, F. A., et al. (2016). "A dataset for breast cancer histopathological image classification." 63(7): 1455-1462.
• Stenkvist, B., et al. (1978). "Computerized nuclear morphometry as an objective method for characterizing human cancer cell populations." 38(12): 4688-4697.
• Wenzhong, L., et al. (2020). "Classifications of Breast Cancer Images by Deep Learning."