Breast Cancer Detection Using Neural Networks

Cancer is the deadliest disease in the world, and it primarily strikes women. As early cancer detection can aid in the disease's treatment, it is imperative that the primary goal be the scientific discovery of a cancer cure. After analyzing 41 publications, we discovered that there are numerous techniques for detecting cancer. We demonstrated a deep learning neural network method for breast cancer diagnosis in this work using the Breast Cancer database. This study illustrates how deep technology and datasets can be used to learn about breast cancer diagnosis. There are three sections to this work. We started by gathering the dataset, dividing it into training and testing sets, applying a pre-processing technique to scale and filter the data, and making some graphs for data visualization. A training dataset was used to test the model, and it achieved an accuracy of 99.67%. We have shown that deep learning technologies can accurately diagnose breast cancer using the Wisconsin Breast Dataset. The dataset comprises 30 features in total, and the database contains 569 rows. In this paper, we use eleven pre-processed features that we obtained to identify breast cancer. But first, we used several pre-processing techniques including Label Encoder, Normalizer, and Standard Scaler, to apply in the model and achieve good accuracy to enhance the model for breast cancer detection.