{"title":"Machine learning and deep learning for breast cancer risk prediction and diagnosis: a Survey","authors":"","doi":"10.46632/daai/2/4/19","DOIUrl":null,"url":null,"abstract":"Breast cancer is the widest spreading disease among women globally. The prevalence rate of breast cancer continued to rise in the last few decades. The mitotic count is a relevant factor for grading invasive breast cancer. Early analysis is an extremely imperative step in treatment. However, it is not an easy one due to several skepticisms in detection which employ mammograms. Since it is subject to human prone error, requires more time for completion and the nuclei look similar during all stages of mitosis, automatic detection of mitosis is a good solution to overcome these problems. Detailed analysis of breast cancer normally requires medical images of different methods. The sensitivity and specificity of the diagnosis largely depend on the experiences of the radiologists, and uncertain diagnosis is quite frequent because of resolution limitations and the concerns of lawsuits arisen from wrong diagnosis or undetected lesions. In this paper, the top methodologies used for mitosis detection are analyzed. There are many algorithms for classification and prediction of breast cancer: Support Vector Machine (SVM), Decision Tree (CART), k Nearest Neighbors (KNN), Random Forest (RF), and Bayesian Networks (BN). The Wisconsin data set was used to\nanalyze breast cancer as a training set to assess and measure the performance of the three ML classifiers in terms of key frameworks such as accuracy, recall, precision, and ROC. The outcome obtained in this paper provides a critique of the stateofart ML techniques for breast cancer detection. It was also found that the ensemble classifier gives better performance. A preliminary experiment conducted on cascaded RF and Artificial Neural Network (ANN) results in better accuracy than individual classifiers. The paper shows how we can use deep learning technology diagnosis of breast cancer using MIAS Dataset. A deep learning approach is almost used for immense task objective Image processing, Computer Vision, Medical Diagnosis, and Neural Language Processing.","PeriodicalId":76806,"journal":{"name":"Vital and health statistics. Ser. 4, Documents and committee reports","volume":"28 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2022-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Vital and health statistics. Ser. 4, Documents and committee reports","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.46632/daai/2/4/19","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
Breast cancer is the widest spreading disease among women globally. The prevalence rate of breast cancer continued to rise in the last few decades. The mitotic count is a relevant factor for grading invasive breast cancer. Early analysis is an extremely imperative step in treatment. However, it is not an easy one due to several skepticisms in detection which employ mammograms. Since it is subject to human prone error, requires more time for completion and the nuclei look similar during all stages of mitosis, automatic detection of mitosis is a good solution to overcome these problems. Detailed analysis of breast cancer normally requires medical images of different methods. The sensitivity and specificity of the diagnosis largely depend on the experiences of the radiologists, and uncertain diagnosis is quite frequent because of resolution limitations and the concerns of lawsuits arisen from wrong diagnosis or undetected lesions. In this paper, the top methodologies used for mitosis detection are analyzed. There are many algorithms for classification and prediction of breast cancer: Support Vector Machine (SVM), Decision Tree (CART), k Nearest Neighbors (KNN), Random Forest (RF), and Bayesian Networks (BN). The Wisconsin data set was used to
analyze breast cancer as a training set to assess and measure the performance of the three ML classifiers in terms of key frameworks such as accuracy, recall, precision, and ROC. The outcome obtained in this paper provides a critique of the stateofart ML techniques for breast cancer detection. It was also found that the ensemble classifier gives better performance. A preliminary experiment conducted on cascaded RF and Artificial Neural Network (ANN) results in better accuracy than individual classifiers. The paper shows how we can use deep learning technology diagnosis of breast cancer using MIAS Dataset. A deep learning approach is almost used for immense task objective Image processing, Computer Vision, Medical Diagnosis, and Neural Language Processing.