An Improved Neural Network Methodology to Predict Breast Cancers in Earlier Stages Using Artificial Intelligence Logic

A large percentage of women on Earth will be touched by breast cancer at some point in their lives. It ranks second in terms of female cancer fatalities. On the other hand, cancer can be cured with early detection and effective treatment. By enabling patients to obtain therapeutic treatment in a timely manner, early identification of breast cancer greatly improves prognosis and survival prospects. In addition, patients can be helped to avoid unnecessary therapy by accurate benign tumor categorization. Breast cancer treatment typically includes radiation therapy, chemotherapy, or a mix of the two. A person's life can be spared if cancer is detected in its early stages. This field is highly dependent on artificial intelligence (AI). Consequently, when it comes to the prediction of breast cancer, there is still a considerable difficulty for both academics as well as medical professionals. This study's objective is to determine the probability that an individual will acquire breast cancer and to measure that probability. This study presents a new approach to breast cancer prediction using AI; the Artificial Intelligence based Breast Cancer Predictor (AIBCP). Cross-validation is performed on AIBCP using a traditional deep learning model known as the Convolutional Neural Network in order to evaluate how well it works. The following steps were taken in order to analyze the dataset, which contains 1059 observations and 28 features: data cleaning, exploratory analysis, training, testing, and validation. Precision, F1 count, specificity, and classification accuracy were the metrics used to assess the models' efficacy. Both the training and the results show that the six models that were trained can produce the best results when it comes to classification and prediction.