Breast Cancer Detection and Classification from Mammogram Images Using Improved Convolutional Neural Network Model

A prominent application in healthcare is the classification and detection of breast cancer using convolutional neural networks (CNNs). It entails creating a system that can precisely pinpoint malignant spots or gauge the tendency of breast cancer through the examination of diagnostic pictures like mammograms. This study focuses on the use of Convolutional Neural Networks for the detection and classification of breast cancer. The proposed model analyzes mammogram images to distinguish cancerous and non-cancerous instances automatically. Preparing the dataset developing the CNNs model’s architecture, training the model, assessing its performance, and producing predictions based on hypothetical data are all the steps in the process. The CNN model learns to extract relevant features and patterns from a large dataset, increasing classification accuracy. Tests show that the CNN model is effective at correctly identifying breast cancer. This method will help medical professionals diagnose patients quickly and correctly, improving patient outcomes in the fight against breast cancer. The prediction can be improved using post-processing techniques and the trained model can be used in practical applications. The model accuracy was 0.98 and the model loss was 0.14.