Published Paper Details:

Breast cancer is the second most common cause of mortality for women globally. However, with early identification and prevention, the risk of dying can be considerably reduced. Breast cells are the site of development for breast cancer, which affects women rather frequently. Breast cancer is the disease that claims the lives of the most women after lung cancer. In existing systems which use algorithms such as Decision Trees, SVM, Random Forest etc. have limitations like scalability, overfitting of the data, feature dependence, computationally intensive etc. which we try to overcome by using Deep Neural Networks like CNN. In order to improve automated breast cancer identification (WSI), we propose a convolutional neural network (CNN) technique in this study by analyzing hostile ductal carcinoma tissue zones in whole-slide pictures. This overview looks at the results of a proposed system for automatically diagnosing breast cancer using several convolutional neural network (CNN) designs and compares them to those achieved using machine learning (ML) techniques. All buildings were built from a large database.Validation tests were conducted in order to provide quantifiable results using the performance parameters of each approach. Functions like VGG16 and ResNet50 are used in this model to improve the accuracy of the model. Both architectures have shown good performance in image classification tasks, including breast cancer detection. In the last layer of architecture Softmax Activation function is used with which we can obtain meaningful class probabilities for breast cancer prediction. This discusses numerous statistics and examines breast cancer datasets to improve the accuracy of breast cancer diagnosis using convolutional neural network techniques. After analysis, the data may be looked at, evaluated, and used for training. Following that, error histograms were extracted from the dataset in order to create the confusion matrix. We may estimate accuracy levels as a result, and we can achieve high accuracy above 90 percent. The outcomes of a breast cancer prediction model can differ based on various factors, including the dataset's quality and size, the selected model architecture, and the employed optimization techniques. Ultimately, the prediction model yields a classification result that indicates whether an individual has breast cancer or not.