Affiliation:
1. Raghu Institute of Technology, Visakhapatnam, India
Abstract
Cancer is a group of diseases that damage tissues by the uncontrolled proliferation of cells. The difficulty of distinguishing skin cancer, which is a common type of cancer, without technical support necessitates studies that can help specialists in the diagnosis phase. In this study, a deep learning model with 7 convolution layers and 3 neural layers was designed to classify the HAM10000 dataset, which consists of 7 classes and includes dermoscopic images. The accuracy rate for the test data of the proposed model was calculated as 99.01%. This result shows that the proposed model can help experts in diagnosing skin cancer.
Reference19 articles.
1. American Cancer Society: Cancer facts and figures 2018. Available: https://www.cancer.org/content/dam/cancer-org/research/cancer-factsand-statistics/annual-cancer-facts-and-figures/2018/cancer-facts-andfigures-2018.pdf , Accessed: 15 Aug 2018.
2. N. C. F. Codella et al., “Deep learning ensembles for melanoma recognition in dermoscopy images” IBM Journal of Research and Development, Vol.61, pp.5:1- 5:15, 2017.
3. N. Codella et al., “Deep learning, sparse coding, and SVM for melanoma recognition in dermoscopy images”, LNCS, Springer, vol. 9352, pp. 118–126, 2015. Available: https://doi.org/10.1007/978-3- 319-24888-2_15 , Accessed: 15 Aug 2018
4. N. K. Mishra and M. E. Celebi. “An overview of melanoma detection in dermoscopy images using image processing and machine learning”, 2016. Available: https://arxiv.org/abs/1601.07843 , Accessed: 15 Aug 2018.
5. Masood, A. & Al-Jumaily, A. A. Computer aided diagnostic support system for skin cancer: a review of techniques and algorithms. Int. J. Biomed. Imaging 2013, vol. 323268, 2013.