Tasneem Alkarakatly, Shatha Eidhah, Miaad Al-Sarawani, A. Al-Sobhi, M. Bilal
{"title":"基于深度卷积神经网络的皮肤病变识别","authors":"Tasneem Alkarakatly, Shatha Eidhah, Miaad Al-Sarawani, A. Al-Sobhi, M. Bilal","doi":"10.1109/AECT47998.2020.9194205","DOIUrl":null,"url":null,"abstract":"Skin cancer is a serious public health problem due to its increasing incidence and subsequent high mortality rate. Deep learning is one of the most important approaches in image analysis used to detect melanoma skin cancer. In this paper, we propose a 5-layer Convolutional Neural Network (CNN) for classifying skin lesions of three categories, including melanoma belonging to deadly skin cancer. The CNN based classifier trained and tested on the PH2 dataset of Dermoscopic images, which is developed for research and benchmarking purposes. The proposed model was evaluated by four well-known performance measures namely, classification accuracy, sensitivity, specificity and area under the curve (AUC). It achieved almost 95% accuracy, 94% sensitivity, 97% specificity, and 100% AUC on the test set. Moreover, in one case of the experiment, the proposed model achieved 100% accuracy.","PeriodicalId":331415,"journal":{"name":"2019 International Conference on Advances in the Emerging Computing Technologies (AECT)","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2020-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"13","resultStr":"{\"title\":\"Skin Lesions Identification Using Deep Convolutional Neural Network\",\"authors\":\"Tasneem Alkarakatly, Shatha Eidhah, Miaad Al-Sarawani, A. Al-Sobhi, M. Bilal\",\"doi\":\"10.1109/AECT47998.2020.9194205\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Skin cancer is a serious public health problem due to its increasing incidence and subsequent high mortality rate. Deep learning is one of the most important approaches in image analysis used to detect melanoma skin cancer. In this paper, we propose a 5-layer Convolutional Neural Network (CNN) for classifying skin lesions of three categories, including melanoma belonging to deadly skin cancer. The CNN based classifier trained and tested on the PH2 dataset of Dermoscopic images, which is developed for research and benchmarking purposes. The proposed model was evaluated by four well-known performance measures namely, classification accuracy, sensitivity, specificity and area under the curve (AUC). It achieved almost 95% accuracy, 94% sensitivity, 97% specificity, and 100% AUC on the test set. Moreover, in one case of the experiment, the proposed model achieved 100% accuracy.\",\"PeriodicalId\":331415,\"journal\":{\"name\":\"2019 International Conference on Advances in the Emerging Computing Technologies (AECT)\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2020-02-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"13\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2019 International Conference on Advances in the Emerging Computing Technologies (AECT)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/AECT47998.2020.9194205\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2019 International Conference on Advances in the Emerging Computing Technologies (AECT)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/AECT47998.2020.9194205","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Skin Lesions Identification Using Deep Convolutional Neural Network
Skin cancer is a serious public health problem due to its increasing incidence and subsequent high mortality rate. Deep learning is one of the most important approaches in image analysis used to detect melanoma skin cancer. In this paper, we propose a 5-layer Convolutional Neural Network (CNN) for classifying skin lesions of three categories, including melanoma belonging to deadly skin cancer. The CNN based classifier trained and tested on the PH2 dataset of Dermoscopic images, which is developed for research and benchmarking purposes. The proposed model was evaluated by four well-known performance measures namely, classification accuracy, sensitivity, specificity and area under the curve (AUC). It achieved almost 95% accuracy, 94% sensitivity, 97% specificity, and 100% AUC on the test set. Moreover, in one case of the experiment, the proposed model achieved 100% accuracy.