{"title":"An AI Based Support System For The Diagnosis Of Breast Cancer","authors":"Debabrata Swain, Utsav Mehta, Ayush Bhatt, Ashwini Ramanuj","doi":"10.1109/iSSSC56467.2022.10051470","DOIUrl":null,"url":null,"abstract":"At present, the health-care system is facing a drastic increase in the number of cancer patients. The huge number of Breast cancer cases among women and its constant increase is a matter of concern for the clinical support systems. Early diagnosis plays a crucial role in the treatment of such a fatal ailment. The late diagnosis is the main cause of the death of many people around the globe. Also, Image processing techniques do exist for the prediction of breast cancer however the scope of misdiagnosis still exists in this technique. For this reason, timely and accurate screening of breast cancer is a major challenge for the clinical support system. In such cases, machine learning can be used as an effective tool to reduce uncertainty in clinical decision-making. Machine learning is the technique for making the machine capable through a large amount of data to perform learning and produce useful outputs. In this work, a machine learning based classifier is developed using the Support vector machine for the diagnosis of breast cancer illness. The clinical data used for the creation and validation of the model is obtained from the UCI repository. SMOTE based oversampling has been performed to balance the classes of malignant and benign tumors in the dataset. A set of seven important features were selected based on their f-value to reduce the time as well as the cost of medical examination. The proposed classifier has reported a testing accuracy of 98.32%.","PeriodicalId":334645,"journal":{"name":"2022 IEEE 2nd International Symposium on Sustainable Energy, Signal Processing and Cyber Security (iSSSC)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2022 IEEE 2nd International Symposium on Sustainable Energy, Signal Processing and Cyber Security (iSSSC)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/iSSSC56467.2022.10051470","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
At present, the health-care system is facing a drastic increase in the number of cancer patients. The huge number of Breast cancer cases among women and its constant increase is a matter of concern for the clinical support systems. Early diagnosis plays a crucial role in the treatment of such a fatal ailment. The late diagnosis is the main cause of the death of many people around the globe. Also, Image processing techniques do exist for the prediction of breast cancer however the scope of misdiagnosis still exists in this technique. For this reason, timely and accurate screening of breast cancer is a major challenge for the clinical support system. In such cases, machine learning can be used as an effective tool to reduce uncertainty in clinical decision-making. Machine learning is the technique for making the machine capable through a large amount of data to perform learning and produce useful outputs. In this work, a machine learning based classifier is developed using the Support vector machine for the diagnosis of breast cancer illness. The clinical data used for the creation and validation of the model is obtained from the UCI repository. SMOTE based oversampling has been performed to balance the classes of malignant and benign tumors in the dataset. A set of seven important features were selected based on their f-value to reduce the time as well as the cost of medical examination. The proposed classifier has reported a testing accuracy of 98.32%.