No abstract

No abstract

Breast Cancer is one of the severe diseases causing large number of deaths in women. So there is a need for efficient technique that diagnoses such cells without the involvement of human, with high accuracies. In the first phase, a complete normalization scheme is introduced to address the problem of color variation in images jointly caused by inconsistent biopsy staining and non-standard imaging condition. This scheme is robust to parameters and insensitive to content of the image and achromatic colors. Then In second phase, a Feed forward back propagation neural network classify benign and malignant tumor and also classify breast cancer tumor in type1, type2 and type3. It can be concluded that it works as promising tool for classification of cancer cells. Breast Cancer is one of the severe disease causing large number of deaths in women. So there is a need for efficient technique that diagnoses such cells without the involvement of human, with high accuracies. In the first phase, a complete normalization scheme is introduced to address the problem of color variation in images jointly caused by inconsistent biopsy staining and non-standard imaging condition. This scheme is robust to parameters and insensitive to content of the image and achromatic colors. Then In second phase, a Feed forward back propagation neural network classify benign and malignant tumor and also classify breast cancer tumor in type1, type2 and type3. It can be concluded that it works as promising tool for classification of cancer cells.