Sai Pranav Kothapalli, Panchumarthi Sri Hari Priya, Vempada Sagar Reddy, Botta Lahya, Prashanth Ragam
{"title":"基于SVM和CNN的黑色素瘤皮肤癌检测","authors":"Sai Pranav Kothapalli, Panchumarthi Sri Hari Priya, Vempada Sagar Reddy, Botta Lahya, Prashanth Ragam","doi":"10.4108/eetpht.9.4340","DOIUrl":null,"url":null,"abstract":"In the field of cancer detection and prevention, doctors and patients are facing numerous challenges when it comes to cancer prediction. Melanoma skin cancer is a deadly type of skin cancer with a multitude of variants spread across the world. Traditional methods involved manual inspection followed by various tests of samples. This time-consuming work and inaccurate predictions sometimes risk the overall health of the patient. The two aspects of solving skin cancer detection problems utilising both conventional image-processing techniques and methods based on machine learning and deep learning are elaborated in this article. It gives a review of current skin cancer detection techniques, weighs the benefits and drawbacks of those techniques, and introduces some relevant cancer datasets. The proposed method focuses mainly on Melanoma skin cancer detection and its previous stages (Common Nevus and Atypical Nevus). The methods being proposed employ a blend of colour, texture, and shape characteristics to derive distinguishing attributes from the images. Using CNN (convolutional neural networks) and SVM (support vector machine) algorithms to identify the type of skin cancer the patient is affected with and achieved an accuracy of 92% and 95% respectively.","PeriodicalId":36936,"journal":{"name":"EAI Endorsed Transactions on Pervasive Health and Technology","volume":" 43","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2023-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Melanoma Skin Cancer Detection using SVM and CNN\",\"authors\":\"Sai Pranav Kothapalli, Panchumarthi Sri Hari Priya, Vempada Sagar Reddy, Botta Lahya, Prashanth Ragam\",\"doi\":\"10.4108/eetpht.9.4340\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"In the field of cancer detection and prevention, doctors and patients are facing numerous challenges when it comes to cancer prediction. Melanoma skin cancer is a deadly type of skin cancer with a multitude of variants spread across the world. Traditional methods involved manual inspection followed by various tests of samples. This time-consuming work and inaccurate predictions sometimes risk the overall health of the patient. The two aspects of solving skin cancer detection problems utilising both conventional image-processing techniques and methods based on machine learning and deep learning are elaborated in this article. It gives a review of current skin cancer detection techniques, weighs the benefits and drawbacks of those techniques, and introduces some relevant cancer datasets. The proposed method focuses mainly on Melanoma skin cancer detection and its previous stages (Common Nevus and Atypical Nevus). The methods being proposed employ a blend of colour, texture, and shape characteristics to derive distinguishing attributes from the images. Using CNN (convolutional neural networks) and SVM (support vector machine) algorithms to identify the type of skin cancer the patient is affected with and achieved an accuracy of 92% and 95% respectively.\",\"PeriodicalId\":36936,\"journal\":{\"name\":\"EAI Endorsed Transactions on Pervasive Health and Technology\",\"volume\":\" 43\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2023-11-09\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"EAI Endorsed Transactions on Pervasive Health and Technology\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.4108/eetpht.9.4340\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"Computer Science\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"EAI Endorsed Transactions on Pervasive Health and Technology","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.4108/eetpht.9.4340","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"Computer Science","Score":null,"Total":0}
In the field of cancer detection and prevention, doctors and patients are facing numerous challenges when it comes to cancer prediction. Melanoma skin cancer is a deadly type of skin cancer with a multitude of variants spread across the world. Traditional methods involved manual inspection followed by various tests of samples. This time-consuming work and inaccurate predictions sometimes risk the overall health of the patient. The two aspects of solving skin cancer detection problems utilising both conventional image-processing techniques and methods based on machine learning and deep learning are elaborated in this article. It gives a review of current skin cancer detection techniques, weighs the benefits and drawbacks of those techniques, and introduces some relevant cancer datasets. The proposed method focuses mainly on Melanoma skin cancer detection and its previous stages (Common Nevus and Atypical Nevus). The methods being proposed employ a blend of colour, texture, and shape characteristics to derive distinguishing attributes from the images. Using CNN (convolutional neural networks) and SVM (support vector machine) algorithms to identify the type of skin cancer the patient is affected with and achieved an accuracy of 92% and 95% respectively.