Breast Cancer Histopathological Image Classification

Yıl 2021, , 87 - 94, 30.01.2021

Ebru Erdem , Tolga Aydin

https://doi.org/10.17671/gazibtd.746673

Cited By: 1

Öz

Breast cancer is one of the most common types of cancer in women worldwide, after lung cancer. Early diagnosis and classification of cancer can positively affect the healing process of patients. In this study, deep learning approaches for cancer detection in chest histopathological images are presented. The success of deep learning architectures may vary depending on the problem. In this study, classification is made using pre-trained CNN architectures, VGG16, Inception-V3, and the network (VGG16 + Inception-V3), which is a combination of two deep neural networks. The concatenate network structure in the study was named as VIHist. The performance of the proposed approaches has been examined on the BreakHist dataset. The images used for detection are 40X magnified image slides. In results, the concatenate network structure (VIHist) gave the highest accuracy rate with 99.03% accuracy. Inception-V3 network showed ~ 6% superior performance than VGG16 deep neural network. Although there is no pathology knowledge on the disease, 98.3% ± 1% accuracy was achieved in the detection of diseases with the proposed deep learning architectures. When the results are examined, it is seen that the performance is higher than the successful studies in the literature.

Anahtar Kelimeler

breast cancer, Inception-V3, VGG16

Göğüs Kanseri Histopatolojik Görüntü Sınıflandırması

Öz

Meme kanseri, dünya genelinde kadınlarda, akciğer kanserinden sonra en çok rastlanan kanser türlerinden biridir. Kanserin erken teşhisi ve sınıflandırılması hastaların iyileşme sürecine olumlu etki edebilmektedir. Bu çalışmada, göğüs histopatolojik görüntülerinde kanser tespiti için derin öğrenme yaklaşımları sunulmuştur. Derin öğrenme mimarilerinin başarısı probleme özgü değişebilmektedir. Bu çalışmada, önceden eğitilmiş CNN mimarilerinden, VGG16, Inception-V3 ve iki derin sinir ağın birleşimi olan ağ (VGG16+Inception-V3) kullanılarak sınıflandırma yapılmıştır. Çalışma içerisinde birleştirme ağ yapısı VIHist olarak adlandırılmıştır. Önerilen yaklaşımların performansı, BreakHist veri seti üzerinde incelenmiştir. Tespit için kullanılan görüntüler 40X büyütülmüş görüntü slaytlarıdır. Elde edilen bulgularda, %99.03 başarı ile birleştirme ağ yapısı (VIHist) en yüksek doğruluk oranını vermiştir. Inception-V3 ağı, VGG16 derin sinir ağına göre ~%6 daha üstün performans göstermiştir. Hastalık üzerinde patoloji bilgisine sahip olunmamasına rağmen, önerilen derin öğrenme mimarileri ile hastalık tespitinde %98.3 ± %1 başarı elde edilmiştir. Sonuçlar incelendiğinde, literatürdeki başarılı çalışmalara göre performansın daha yüksek bulunduğu görülmüştür.

Anahtar Kelimeler

Göğüs kanseri, Inception-V3, VGG16

Kaynakça

• M. M. Saritas, A. Yasar, “Performance Analysis of ANN and Naive Bayes Classification Algorithm for Data Classification”, International Journal of Intelligent Systems and Applications in Engineering, 7(2), 88-91, 2019.
• D. Bardou, K. Zhang, S.M. Ahmad, “Classification of breast cancer based on histology images using convolutional neural networks”, IEEE Access, 6, 24680-24693, 2018.
• M. Amrane, et al., “Breast cancer classification using machine learning”, Electric Electronics, Computer Science, Biomedical Engineerings' Meeting (EBBT), Istanbul, Turkey, 17885967, 2018.
• Z. Han, et al., “Breast cancer multi-classification from histopathological images with structured deep learning model", Scientific reports, 7(1), 1-10, 2017.
• Y. Benhammou, et al., “BreakHis based breast cancer automatic diagnosis using deep learning: Taxonomy, survey and insights”, Neurocomputing, 375, 9-24, 2020.
• A.-A. Nahid, A. Mikaelian, Y. Kong, “Histopathological breast-image classification with restricted Boltzmann machine along with backpropagation”, Biomedical Reseacrh, 29(10), 2018.
• S.A. Adeshina, et al., “Breast cancer histopathology image classification with deep convolutional neural networks”, 14th International Conference on Electronics Computer and Computation (ICECCO), Kaskelen, Kazakhstan, 18434739, 29 November-1 December, 2018.
• N. Bayramoglu, J. Kannala, and J. Heikkilä. “Deep learning for magnification independent breast cancer histopathology image classification”, 23rd International conference on pattern recognition (ICPR), Cancun, Mexico, 16824576, 4-8 December, 2016.
• F. F. Ting, Y.J. Tan, K.S. Sim, “Convolutional neural network improvement for breast cancer classification”, Expert Systems with Applications, 120, 103-115, 2019.
• P. Sudharshan, et al., “Multiple instance learning for histopathological breast cancer image classification”, Expert Systems with Applications, 117, 103-111, 2019.
• M. Z. Alom, et al., “Breast cancer classification from histopathological images with inception recurrent residual convolutional neural network”, Journal of digital imaging, 32(4), 605-617, 2019.
• S. Khan, et al., “A novel deep learning based framework for the detection and classification of breast cancer using transfer learning”, Pattern Recognition Letters, 125, 1-6, 2019.
• M. Saini, et al. “Data Augmentation of Minority Class with Transfer Learning for Classification of Imbalanced Breast Cancer Dataset Using Inception-V3”, Iberian Conference on Pattern Recognition and Image Analysis, 409-420, 2019.
• J. Chang, et al. “A method for classifying medical images using transfer learning: A pilot study on histopathology of breast cancer ”, 2017 IEEE 19th International Conference on e-Health Networking, Applications and Services (Healthcom), 1-4, 2017.
• M. Gour, et al., “Residual learning based CNN for breast cancer histopathological image classification”, International Journal of Imaging Systems and Technology, 30(3), 621-635, 2020.
• S. Sharma, et al., “Conventional Machine Learning and Deep Learning Approach for Multi-Classification of Breast Cancer Histopathology Images—a Comparative Insight”, Journal of Digital Imaging, 33, 632-654, 2020.
• R. Man, et al., “Classification of Breast Cancer Histopathological Images Using Discriminative Patches Screened by Generative Adversarial Networks”, IEEE Access, 8, 2020.
• C. Szegedy, et al. “Rethinking the inception architecture for computer vision”, Proceedings of the IEEE conference on computer vision and pattern recognition, 2818-2826, 2016.
• K. Simonyan, and A. Zisserman, “Very deep convolutional networks for large-scale image recognition”, arXiv preprint arXiv:1409.1556, 2014.
• Internet: Sitapatology, P.A.a. Excelência no Diagnóstico. Apoio ao cliente, http://www.prevencaoediagnose.com.br/, 12.03.2020.
• F. Chollet, “Keras”, 2015.
• A. Krizhevsky, I. Sutskever, and G.E. Hinton. “Imagenet classification with deep convolutional neural networks”, Advances in neural information processing systems, 2012.
• C. Szegedy, et al., “Going deeper with convolutions”, Proceedings of the IEEE conference on computer vision and pattern recognition, 1-9, 2015.