Breast Cancer Detection Using Machine Learning Algorithms
Nandini S R, Jnanesh Gowda K S1, Bharath S R2, Kishor L D3, Manjunath B S4
Nandini S R, Assistant professor, BGS Institute of Technology
1Jnanesh Gowda K S, Department of Computer Science and Engineering, BGS Institute of Technology
2Bharath S R, Department of Computer Science and Engineering, BGS Institute of Technology
3Kishor L D, Department of Computer Science and Engineering, BGS Institute of Technology
4Manjunath B S, Department of Computer Science and Engineering, BGS Institute of Technology
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract – The rapid development of in-depth learning, a family of machine learning techniques, has generated a great deal of interest in its application to the problems of medical thinking. Here, we developed an in-depth study algorithm that can accurately detect breast cancer in mammograms testing using a “end-to-end” training method that effectively uses training data sets with complete clinical annotations or only cancer status (label) for the whole picture. In this method, wound annotations are only required for the first stage of training, and subsequent stages require image-level labels only, which removes reliance on wound annotations that are not readily available. Our entire convolutional network method of separating test mammograms has found much better performance compared to previous methods. The rapid development of in-depth learning, a family of machine learning techniques, has generated a great deal of interest in its application to the problems of medical thinking. Here, we developed an in-depth study algorithm that can accurately detect breast cancer in mammograms testing using a “end-to-end” training method that effectively uses training data sets with complete clinical annotations or only cancer status (label) for the whole picture. In this method, wound annotations are only required for the first stage of training, and subsequent stages require image-level labels only, which removes reliance on wound annotations that are not readily available. Our entire convolutional network method of separating test mammograms has found much better performance compared to previous methods.
Key Words: Machine learning, Mammograms