Hybrid Convolutional Neural Network Architectures for Skin Cancer Classification

Yıl 2021, Sayı: 28, 694 - 701, 30.11.2021

Emine Cengil Ahmet Çınar Muhammed Yıldırım

https://doi.org/10.31590/ejosat.1010266

Cited By: 3

Öz

Skin cancer is a common form of cancer seen in humans. Like other diseases, early diagnosis of skin cancer is vital. In the study, deep learning architectures, which are popular machine learning algorithms, are used to classify skin cancer. In order to increase accuracy performance, hybrid structures are realized using K-Nearest neighbor (KNN), Support vector machine (SVM) and Decision tree (DT). After feature extraction using convolutional neural network, KNN, SVM and DT are applied separately for classification. While the KNN and SVM of the produced hybrid structures increase performance, the use of the decision tree has negatively affected the performance. After the training and validation processes with the seven-class skin cancer mnist: ham10000 dataset containing dermatological images, the validation accuracy and confusion matrix criteria of the architectures are reported. Eight different architectures are implemented. The highest accuracy is provided by the structure in which the last layer of Alexnet architecture is replaced by the SVM classifier.

Anahtar Kelimeler

Skin Cancer, Classification, CNN, Deep Learning, Hybrid Structures, KNN, SVM, DT

Cilt Kanseri Sınıflandırması için Hibrit Evrişimli Sinir Ağı Mimarileri

Öz

Deri kanseri, insanlarda görülen yaygın bir kanser türüdür. Diğer hastalıklarda olduğu gibi cilt kanserinde de erken teşhis hayati önem taşımaktadır. Çalışmada cilt kanserini sınıflandırmak için popüler makine öğrenmesi algoritmaları olan derin öğrenme mimarileri kullanılmaktadır. Doğruluk performansını artırmak için K-En yakın komşu (KNN), Destek vektör makinesi (SVM) ve Karar ağacı (DT) kullanılarak hibrit yapılar gerçekleştirilmektedir. Evrişimli sinir ağı kullanılarak öznitelik çıkarıldıktan sonra, sınıflandırma için KNN, SVM ve DT ayrı ayrı uygulanır. Üretilen hibrit yapıların KNN ve SVM'si performansı artırırken, karar ağacının kullanılması performansı olumsuz etkilemektedir. Dermatolojik görüntüleri içeren yedi sınıflı cilt kanseri mnist:ham10000 veri seti ile yapılan eğitim ve doğrulama işlemlerinden sonra mimarilerin doğrulama doğruluğu ve karmaışıklık matrisi kriterleri raporlanır. Sekiz farklı mimari uygulanmaktadır. En yüksek doğruluk, Alexnet mimarisinin son katmanının SVM sınıflandırıcısı ile değiştirildiği yapı tarafından sağlanmaktadır.

Anahtar Kelimeler

Deri Kanseri, Sınıflandırma, ESA, Derin Öğrenme, Hibrit Yapılar, KNN, SVM, DT.

Kaynakça

• Öztürk, Ş., & Özkaya, U. (2020). Skin lesion segmentation with improved convolutional neural network. Journal of digital imaging, 33(4), 958-970.
• Yıldız, O. (2019). Melanoma detection from dermoscopy images with deep learning methods: Acomprehensive study. Journal of the Faculty of Engineering and Architecture of Gazi University, 34(4), 2241-2260.
• Chang, H. (2017). Skin cancer reorganization and classification with deep neural network. arXiv preprint arXiv:1703.00534.Bejan, A. (2015). Constructal thermodynamics. Constructal Law & Second Law Conference, Parma, pp. S1-S8.
• Ünlü, E. I., Çınar, A. (2018). Classification of skin images with respect to melanoma and nonmelanoma using the deep neural network.
• Codella, N., Cai, J., Abedini, M., Garnavi, R., Halpern, A., & Smith, J. R. (2015). Deep learning, sparse coding, and SVM for melanoma recognition in dermoscopy images. In International workshop on machine learning in medical imaging (pp. 118-126). Springer, Cham.
• YILDIRIM, M., & ÇINAR, A. (2021). Classification of Skin Cancer Images with Convolutional Neural Network Architectures. Turkish Journal of Science and Technology, 16(2), 187-195.
• Purnama, I. K. E., Hernanda, A. K., Ratna, A. A. P., Nurtanio, I., Hidayati, A. N., Purnomo, M. H., ... & Rachmadi, R. F. (2019). Disease Classification based on Dermoscopic Skin Images Using Convolutional Neural Network in Teledermatology System. In 2019 International Conference on Computer Engineering, Network, and Intelligent Multimedia (CENIM) (pp. 1-5). IEEE.
• Pai, K., & Giridharan, A. (2019, October). Convolutional Neural Networks for classifying skin lesions. In TENCON 2019-2019 IEEE Region 10 Conference (TENCON) (pp. 1794-1796). IEEE.
• Tschandl, P., Rosendahl, C., & Kittler, H. (2018). The HAM10000 dataset, a large collection of multi-source dermatoscopic images of common pigmented skin lesions. Scientific data, 5(1), 1-9.
• O'Shea, K., & Nash, R. (2015). An introduction to convolutional neural networks. arXiv preprint arXiv:1511.08458.
• Wolfe, J., Jin, X., Bahr, T., & Holzer, N. (2017). Application of softmax regression and its validation for spectral-based land cover mapping. The International Archives of Photogrammetry, Remote Sensing and Spatial Information Sciences, 42, 455.
• Maimon, O., & Rokach, L. (Eds.). (2005). Data mining and knowledge discovery handbook.
• Papernot, N., & McDaniel, P. (2018). Deep k-nearest neighbors: Towards confident, interpretable and robust deep learning. arXiv preprint arXiv:1803.04765.
• Tang, Y. (2013). Deep learning using linear support vector machines. arXiv preprint arXiv:1306.0239.
• Cengil, E., & Cinar, A. (2018, September). A deep learning based approach to lung cancer identification. In 2018 International Conference on Artificial Intelligence and Data Processing (IDAP) (pp. 1-5). IEEE.
• Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2012). Imagenet classification with deep convolutional neural networks. Advances in neural information processing systems, 25, 1097-1105.
• He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 770-778).
• Murugan, S., & Verwillow, A. DeepDerm: Detection of Cancerous Skin Lesions Through Deep Learning.