Classification of Dermoscopy Images with Feed Forward Neural Network, Decision Trees and Random Forest

Abstract

Today, cancer diseases are increasing rapidly. Although skin cancer is less common in populations than other types of cancer, it is a cancer type with a high lethality in late diagnosis. Just as harmful rays from the sun can trigger skin cancer, genetic factors are also a major factor in the formation of skin cancer. In skin cancer, the mortality rate is low in early detection, while the survival rate is low in late diagnosis. Classification of malignant (malignant) and benign (benign) lesions from dermoscopy images by using artificial neural networks is thought to facilitate early diagnosis. In this study, the data were taken from the International Collaboration on Skin Imaging (ISIC) data set using the ready data set. After preprocessing was applied to dermoscopy images, entropy, standard deviation, area, homogeneity, contrast, correlation, energy, skewness and kurtosis were extracted in MATLAB. Along with these features, age and gender information, which are demographic information, are also added to the features to be used for classification. These features are classified using MATLAB and WEKA programs. It is classified by feedforward neural network and decision trees algorithm in MATLAB, it is classified using WEKA program for random forest algorithm. The results were obtained by training the networks with these three methods.

Keywords

• Skin cancer (melanoma)
• melanocytic nevus
• Neural Network
• classification
• decision trees
• random forest

References

1. (2020). Medicalpark. [Online]. Avaible: https://www.medicalpark.com.tr/cilt-kanseri/hg-1808
2. Yavuz, G. Ö., & Yavuz, İ. H. (2014). Melanositik Nevusler. Van Tıp Dergisi, 21(4), 259-268.
3. “Cancer facts and figures 2016,” American Cancer Society
4. ÖZTÜRK, Banu, et al. "Kutanöz malign melanomda adjuvan medikal tedavi yakla¸sımları." Türk Onkoloji Dergisi 25 (2010): 170-80
5. Codella, N., Cai, J., Abedini, M., Garnavi, R., Halpern, A., & Smith, J. R. (2015, October). 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.
6. Milton, M. A. A. (2019). Automated skin lesion classification using ensemble of deep neural networks in ISIC 2018: Skin lesion analysis towards melanoma detection challenge. arXiv preprint arXiv:1901.10802.
7. Dubal, P., Bhatt, S., Joglekar, C., & Patil, S. (2017, November). Skin cancer detection and classification. In 2017 6th international conference on electrical engineering and informatics (ICEEI) (pp. 1-6). IEEE.
8. Dascalu, A., & David, E. O. (2019). Skin cancer detection by deep learning and sound analysis algorithms: A prospective clinical study of an elementary dermoscope. EBioMedicine, 43, 107-113.

9. Esteva, A., Kuprel, B., Novoa, R. A., Ko, J., Swetter, S. M., Blau, H. M., & Thrun, S. (2017). Dermatologist-level classification of skin cancer with deep neural networks. nature, 542(7639), 115-118.
10. Dorj, U. O., Lee, K. K., Choi, J. Y., & Lee, M. (2018). The skin cancer classification using deep convolutional neural network. Multimedia Tools and Applications, 77(8), 9909-9924.
11. Noel Codella, Veronica Rotemberg, Philipp Tschandl, M. Emre Celebi, Stephen Dusza, David Gutman, Brian Helba, Aadi Kalloo, Konstantinos Liopyris, Michael Marchetti, Harald Kittler, Allan Halpern: “Skin Lesion Analysis Toward Melanoma Detection 2018: A Challenge Hosted by the International Skin Imaging Collaboration (ISIC)”, 2018; https://arxiv.org/abs/1902.03368
12. (2021). Mathworks website [Online]. Avaible: https://www.mathworks.com/help/deeplearning/gs/classify-patterns-with-a-neural-network.html
13. DUYGU, B., KOCAOĞLU, M., & COŞKUN, A. Karar Ağaçları ile Otistik Spektrum Bozukluğu Tanısı Koyma.
14. Chen, J., Li, K., Tang, Z., Bilal, K., Yu, S., Weng, C., & Li, K. (2016). A parallel random forest algorithm for big data in a spark cloud computing environment. IEEE Transactions on Parallel and Distributed Systems, 28(4), 919-933.
15. Roslin, S. E. (2020). Classification of melanoma from Dermoscopic data using machine learning techniques. Multimedia tools and applications, 79(5), 3713-3728.