Detection of Benign and Malignant Skin Cancer from Dermoscopic Images using Modified Deep Residual Learning Model

Abstract

Skin cancer is caused by the uncontrolled proliferation of cells on the skin surface due to damaged DNA structures of them and is among the most common cancer types in the world. If malignant skin cancer is not detected early, it can result in death. For this reason, early and high accuracy detection of skin cancer is important in terms of increasing the chance of survival of patients. In this study, ResNet101 architecture, which is one of the deep residual learning models, is suggested for the detection of malignant skin cancer from dermoscopy images. The model was trained and tested on a dataset of 3297 dermoscopic images from the ISIC 2017 archive. As a result of 10 experiments, average 90,67% and maximum 91,36% accuracy values were reached. In this study, a better performance was obtained compared to previous studies using the same dataset in the literature. In conclusion, the proposed approach has promise in the field of medicine and can help dermatologists diagnose skin cancer.

Keywords

• skin cancer
• classification
• deep residual learning
• pattern recognition
• machine learning

References

1. Saba, T., Khan, M.A., Rehman, A. & Marie-Sainte, S.L. (2019). Region extraction and classification of skin cancer: A heterogeneous framework of deep CNN features fusion and reduction. J. Med. Syst., 43(9), 289.
2. Afza, F., Sharif, M., Khan, M.A., Tariq, U., Yong, H.-S. & Cha, J. (2022). Multiclass skin lesion classification using hybrid deep features selection and extreme learning machine. Sensors, 22(3), 799. doi: 10.3390/s22030799
3. Nadipineni, H. (2020). Method to classify skin lesions using dermoscopic images. arXiv 2008.09418.
4. Bafounta, M.L., Beauchet, A., Aegerter, P. & Saiag, P. (2001). Is dermoscopy (epiluminescence microscopy) useful for the diagnosis of melanoma? Results of a meta-analysis using techniques adapted to the evaluation of diagnostic tests. Arch Dermatol., 137(10), 1343-1350. doi: 10.1001/archderm.137.10.1343
5. Kassem, M.A., Hosny, K.M. & Fouad, M.M. (2020). Skin lesions classification into eight classes for ISIC 2019 using deep convolutional neural network and transfer learning. In IEEE Access, 8, 114822-114832. doi: 10.1109/ACCESS.2020.3003890
6. Sun, Q., Huang, C., Chen, M., Xu, H. & Yang, Y. (2021). Skin lesion classification using additional patient information. BioMed Research International, vol. 2021, Article ID 6673852, 6 pages. doi: 10.1155/2021/6673852
7. Nusraddinov, T. (2018). Skin lesion classification using deep convolutional neural network and HSV color space. Master Thesis, İstanbul Technical University.
8. Yilmaz, E. & Trocan, M. (2021). A modified version of GoogLeNet for melanoma diagnosis. Journal of Information and Telecommunication, 5(3), 395-405. doi: 10.1080/24751839.2021.1893495

9. Romero Lopez, A., Giro-i-Nieto, X., Burdick, J. & Marques, O. (2017). Skin lesion classification from dermoscopic images using deep learning techniques. 13th IASTED International Conference on Biomedical Engineering (BioMed), pp. 49-54. doi: 10.2316/P.2017.852-053.
10. Kaur, R., GholamHosseini, H., Sinha, R. & Lindén, M. (2022). Melanoma classification using a novel deep convolutional neural network with dermoscopic images. Sensors, 22(3), 1134. doi: 10.3390/s22031134
11. Ergün, E. & Kılıç, K. (2021). Derin öğrenme ile artırılmış görüntü seti üzerinden cilt kanseri tespiti. Black Sea Journal of Engineering and Science, 4(4), 192-200. doi: 10.34248/bsengineering.938520
12. Öztürk, E. & İçer, S. (2021). Classification of dermoscopy images with feed forward neural network, decision trees and random forest. International Journal of Multidisciplinary Studies and Innovative Technologies, 5(2), 129-135.
13. Ameri, A. (2020). A deep learning approach to skin cancer detection in dermoscopy images. Journal of Biomedical Physics & Engineering, 10(6), 801-806. doi: 10.31661/jbpe.v0i0.2004-1107
14. Demir, F. (2021). Derin öğrenme tabanlı yaklaşımla kötü huylu deri kanserinin dermatoskopik görüntülerden saptanması. Fırat Üniversitesi Mühendislik Bilimleri Dergisi, 33(2), 617-624. doi: 10.35234/fumbd.900170
15. Farooq, M.A., Khatoon, A., Varkarakis, V. & Corcoran, P.M. (2019). Advanced deep learning methodologies for skin cancer classification in prodromal stages. AICS.
16. Soylu, E. & Demir, R. (2021). Development and comparison of skin cancer diagnosis models. Avrupa Bilim ve Teknoloji Dergisi, (28), 1217-1221. doi: 10.31590/ejosat.1013910
17. Codella, N.C., Gutman, D., Celebi, M.E., Helba, B., Marchetti, M.A., Dusza, S.W., Kalloo, A., Liopyris, K., Mishra, N., Kittler, H. et al. (2018). 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), in: Biomedical Imaging (ISBI 2018), IEEE 15th International Symposium on, IEEE, pp. 168–172.
18. 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.
19. Fanconi, C. (2022). Skin cancer: malignant vs. benign. https://www.kaggle.com/fanconic/skin-cancer-malignant-vs-benign, (Last access date: May 20, 2022).
20. LeCun, Y., Boser, B., Denker, J.S., Henderson, D., Howard, R.E., Hubbard, W. & Jackel, L.D. (1989). Backpropagation applied to handwritten zip code recognition. Neural Comput., 1(4), 541-551. doi: 10.1162/neco.1989.1.4.541
21. Goodfellow, I., Bengio, Y. & Courville, A. (2016). Deep Learning. MIT Press.
22. He, K., Zhang, X., Ren, S. & Sun, J. (2016). Deep residual learning for image recognition. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 770-778. doi: 10.1109/CVPR.2016.90.