Dermoscopic classification using Image processing and CNN
Keywords:
melanoma, convolutional neural network, image processing, SE-ResNetAbstract
Melanoma is one of the deadliest forms of skin cancer accounting to more than 70 percent of skin cancer deaths despite it being the least common skin cancer. In the early stages, melanoma can be treated successfully with surgery alone and survival rates are high, but after metastasis survival rates drop significantly as it is known to spreads to other parts of the body. Hence to combat this, Early detection of melanoma is imperative. I propose an automated melanoma detection model by analysis of skin lesion images using SE-ResNeXt a variant of ResNet that uses squeeze-and-excitation blocks to bring significant improvement in performance of existing CNNs. I performed the evaluation using a large publicly available dataset ISIC 2020 Challenge Dataset, generated by the International Skin Imaging Collaboration containing skin lesion images from several primary medical sources, have successfully demonstrated classification performance with an accuracy achieved about 90%.
References
Labani, S., Asthana, S., Rathore, K., & Sardana, K. (2021). Incidence of melanoma and nonmelanoma skin cancers in Indian and the global regions. Journal of Cancer Research and Therapeutics, 17(4), 906.
https://mapmelanoma.org/american-cancer-society-releases-2021-cancer-facts-figures-report
Zito, Patrick M., and Richard Scharf. “Melanoma of the head and neck.” In StatPearls [Internet]. StatPearls Publishing, 2021.
Sankeeth, Revata Sai, et al. “Melanoma disease detection using convolutional neural networks.” 2020 4th International Conference on Intelligent Computing and Control Systems (ICICCS). IEEE, 2020.
Zhang, Long, Hong Jie Gao, Jianhua Zhang, and Bengamin Hasani. “Optimization of the Convolutional Neural Networks for Automatic Detection of Skin Cancer.” Open Medicine 15, 1 (2020): 27-37.
Agero, G.S., Dover J.P. “Dermatoscopy of pigmented skin lesions: A valuable tool for early diagnosis of melanoma. Lancet Oncol. 2001 2(7):443-449.
N. Tajbakhsh, J.Y. Shin, S.R. Gurudu, R.T. Hurst, C.B. Kendall, M.B. Gotway, J. Liang. “Convolutional neural networks for medical image analysis: Full training or fine tuning?” IEEE Trans. Med. Imaging 35 (5) (2016) 1299-1312.
Gouda, Niharika, and J. Amudha. “Skin cancer classification using resnet.” 2020 IEEE 5th International Conference on Computing Communication and Automation (ICCCA). IEEE, 2020.
Maglogiannis, Ilias, and Kostantinos Doukas. “Hair removal on dermoscopy images.” 2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC). IEEE, 2015.
Krizhevsky, Alex, Ilya Sutskever, and Geoffrey E. Hinton. “Imagenet classification with deep convolutional neural networks.” Advances in neural information processing systems 25 (2012).
Skin Cancer Facts & Statistics. Skincancer.org
Risks and causes of melanoma, Cancersresearchuk.org
Why Melanoma Rates Are Increasing in Adolescents and Young Adults, Especially Among Females, Demytra Mitsis, MD, and Nikhil I. Khushalani, MD
Skin cancer statistics. wcrf.org
McMahon, Devon E., et al. “Skin Biopsy Equipment Availability Across 7 Low-Income Countries: A Cross-Sectional Study of 603 Health Facilities.” JAMA dermatology 157.4 (2021): 462-464.
Weinstock, M. A., et al. “Skin biopsy utilization and melanoma incidence among Medicare beneficiaries.” British Journal of Dermatology 176.4 (2017): 949-954.
4 ways to check for skin cancer with your smartphone, Amanda Capritto
Kasmi, Reda, and Karim Mokrani. “Classification of malignant melanoma and benign skin lesions: implementation of automatic ABCD rule.” IET Image Processing 10.6 (2016): 448-455.
Hu Jie, Li Shen, and Gang Sun. “Squeeze-and-excitation networks.” Proceedings of the IEEE conference on computer vision and pattern recognition. 2018.
Downloads
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.