Cilt Kanseri Tanı Modellerinin Geliştirilmesi ve Karşılaştırılması

Öz

Cilt kanseri, cildin en dış tabakası olan epidermisteki anormal hücrelerin kontrolsüz büyümesidir. Anormal hücrelerin hızlı büyümesi ve çoğalması, cildin kötü huylu tümörlerini oluşturur. Araştırmacılar, cilt görüntülerinin bilgisayar analizi ile cilt lekesinin iyi huylu veya kötü huylu olup olmadığını ayırt etmeye çalışırlar. Bu çalışmada malign cilt görüntülerinin bilgisayar analizi ile teşhis edilmesi amaçlanmıştır. Derideki lekeli görünüm, derin transfer öğrenme teknikleri kullanılarak iyi huylu veya kötü huylu olarak sınıflandırılır. Ağ eğitiminde iyi huylu veya kötü huylu cilt lekesi görüntü verileri kullanıldı. Görüntü sınıflandırmasında, Matlab derin öğrenme araç kutusunda bulunan darkNet-19, darkNet-53, squeezeNet, shufleNet mimarileri karşılaştırılmıştır. Yüksek doğrulukta sonuçlar elde edilmiştir. DarkNet-19 mimarisi ile en yüksek performans %89,89 ile elde edilmiştir. DarkNet-53, shuffleNet, pinchNet mimarilerinin performansları sırasıyla %87,36, %86,15, %84,23'tür.

Anahtar Kelimeler

• Cilt kanseri
• Görüntü sınıflandırması
• Derin öğrenme

Development and Comparison of Skin Cancer Diagnosis Models

Abstract

Skin cancer is the uncontrolled growth of abnormal cells in the epidermis, the outermost layer of skin. The rapid growth and proliferation of abnormal cells creates malignant tumors of the skin. With the computer analysis of skin images, researchers are made to distinguish whether the skin spot is benign or malign It is automatically possible to classify whether a skin spot is benign or malignant by computer analysis of skin images. In this study, it was aimed to diagnose malignant skin images by computer analysis. The stained appearance on the skin is classified as benign or malignant using deep transfer learning techniques. Benign or malignant skin spot image data were used in network training. In image classification, darkNet-19, darkNet-53, squeezeNet, shufleNet architectures available in the Matlab deep learning toolbox were compared. High accuracy results have been obtained. The highest performance was achieved with the rate of 89.89% with darkNet-19 architecture. The performances of the networks darkNet-53, shuffleNet, squeezeNet architectures are 87.36%, 86.15%, 84.23% respectively.

Keywords

• Skin cancer
• Image classification
• Deep learning

Kaynakça

1. Rognoni, E., & Watt, F. M. (2018). Skin cell heterogeneity in development, wound healing, and cancer. Trends in cell biology, 28(9), 709-722.
2. Fitzmaurice C, Akinyemiju TF, Al Lami FH, Alam T, Alizadeh-Navaei R, Allen C, et al. (2018)Global, regional, and national cancer incidence, mortality, years of life lost, years lived with disability, and disability-adjusted life-years for 29 cancer groups, 1990 to 2016: a systematic analysis for the global burden of disease study. JAMA Oncol.;4(11):1553–68.
3. Dinehart, S. M. (2000). The treatment of actinic keratoses. Journal of the American Academy of Dermatology, 42(1), S25-S28.
4. Skin Cancer Facts & Statistics [Internet]. 2021. Available from: https://www.skincancer.org/skin-cancer-information/skin-cancer-facts/
5. Murugan, A., Nair, S. A. H., Preethi, A. A. P., & Kumar, K. S. (2021). Diagnosis of skin cancer using machine learning techniques. Microprocessors and Microsystems, 81, 103727.
6. Ogden E, Schofield J. (2013)Benign skin lesions. Medicine (Baltimore).;41(7):406–8.
7. Andrew, T. W., Alrawi, M., & Lovat, P. (2021). Reduction in skin cancer diagnoses in the UK during the COVID‐19 pandemic. Clinical and Experimental Dermatology, 46(1), 145-146.
8. Linares MA, Zakaria A, Nizran P. (2015) Skin Cancer. Prim CareClinics Off Pract.;42(4):645–59.

9. Eigentler, T. K., Leiter, U., Häfner, H. M., Garbe, C., Röcken, M., & Breuninger, H. (2017). Survival of patients with cutaneous squamous cell carcinoma: results of a prospective cohort study. Journal of Investigative Dermatology, 137(11), 2309-2315.
10. Crowson, A. N. (2006). Basal cell carcinoma: biology, morphology and clinical implications. Modern pathology, 19(2), S127-S147.
11. Goyal, M., Knackstedt, T., Yan, S., & Hassanpour, S. (2020). Artificial intelligence-based image classification for diagnosis of skin cancer: Challenges and opportunities. Computers in Biology and Medicine, 104065.
12. Özdemir F. (2007) Diagnosis of Melanoma. TURKDERM [Internet].;41(0):6–14. Available from: https://dx.doi.org/
13. Pacheco, A. G., & Krohling, R. A. (2020). The impact of patient clinical information on automated skin cancer detection. Computers in biology and medicine, 116, 103545.
14. Dorrell, D. N., & Strowd, L. C. (2019). Skin cancer detection technology. Dermatologic clinics, 37(4), 527-536.
15. Holte, K., & Biswas, A. (2017). Pathology of malignant skin tumours. Surgery (Oxford), 35(9), 478-483.
16. Diepgen, T. L., & Mahler, V. (2002). The epidemiology of skin cancer. British Journal of Dermatology, 146, 1-6.
17. ŞENTÜRK, A., & ŞENTÜRK, Z. K. (2016). Yapay Sinir Ağları İle Göğüs Kanseri Tahmini. El-Cezeri Journal of Science and Engineering, 3(2).
18. Sivari E, Civelek Z, Kahraman G. (2020)Artificial neural network model estimating the initial dose of propofol used in general anesthesia. El-Cezeri J Sci Eng.;7(3):1482–95.
19. Umut, K., Yılmaz, A., & Dikmen, Y. (2019). Sağlık alanında kullanılan derin öğrenme yöntemleri. Avrupa Bilim ve Teknoloji Dergisi, (16), 792-808.
20. Lin, T. C., & Lee, H. C. (2020, December). Skin Cancer Dermoscopy Images Classification with Meta Data via Deep Learning Ensemble. In 2020 International Computer Symposium (ICS) (pp. 237-241). IEEE.
21. Mijwil, M. M. (2021). Skin cancer disease images classification using deep learning solutions. Multimedia Tools and Applications, 1-17.
22. Layode, O., Alam, T., & Rahman, M. M. (2019, October). Deep Learning Based Integrated Classification and Image Retrieval System for Early Skin Cancer Detection. In 2019 IEEE Applied Imagery Pattern Recognition Workshop (AIPR) (pp. 1-7). IEEE.
23. Çevik, E., & Zengin, K. (2019). Classification of skin lesions in dermatoscopic images with deep convolution network. Avrupa Bilim ve Teknoloji Dergisi, 309-318.
24. Codella NCF, Gutman D, Celebi ME, Helba B, Marchetti MA, Dusza SW, et al. Skin lesion analysis toward melanoma detection: A challenge at the 2017 International symposium on biomedical imaging (ISBI), hosted by the international skin imaging collaboration (ISIC). Proc - Int Symp Biomed Imaging. 2018;2018-April(Isbi):168–72.
25. Skin Cancer: Malignant vs Benign [Internet]. Available from: https://www.kaggle.com/fanconic/skin-cancer-malignant-vs-benign
26. Redmon, J., & Farhadi, A. (2017). YOLO9000: better, faster, stronger. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 7263-7271).
27. Redmon, J., & Farhadi, A. (2018). Yolov3: An incremental improvement. arXiv preprint arXiv:1804.02767.
28. Iandola, F. N., Han, S., Moskewicz, M. W., Ashraf, K., Dally, W. J., & Keutzer, K. (2016). SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and< 0.5 MB model size. arXiv preprint arXiv:1602.07360.
29. Zhang, X., Zhou, X., Lin, M., & Sun, J. (2018). Shufflenet: An extremely efficient convolutional neural network for mobile devices. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 6848-6856).