In the field of data mining and computational intelligence, many researchers have developed a lot of feasible solutions for the breast cancer detection and classification. Breast Cancer is a malignant tumour that develops from the cells of the breast. In underdeveloped, developing and developed nations, the major cause of death among women is due to breast cancer. To mitigate the death caused by breast cancer, it is essential to detect it much earlier to provide the best treatment. An accurate diagnosis procedure is required which must also be quite reliable so that physicians are able to distinguish from the benign breast tumour from malignant ones. An important real world medical problem is to diagnose the breast cancer in an automated fashion. In this paper, an Expectation Maximization (EM) Based Logistic Regression (LR) is used to detect and classify the breast cancer at an early stage. From the cancer centre of Kuppuswamy Naidu Memorial Hospital, Coimbatore, India, the data was collected and it was classified with EM based LR The stages of cancer are divided into pathological and clinical stages and so the Tumour Node Metasis (TNM) prognostic tools are identified. The various TNM values are given as input variables for the classification of breast cancer. Results show that an average classification accuracy of 95.90% is obtained along with an average performance index of 90.72%.
Expectation maximization based logistic regression for breast cancer classification
2017-04-01
241152 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Dense Photometric Stereo by Expectation Maximization
British Library Conference Proceedings | 2006
|TLE outlier detection based on expectation maximization algorithm
Elsevier | 2021
|