CNN-Based Breast Cancer Ultrasound Medical Image Recognition

This paper proposes a novel data model for the recognition of medical images in breast cancer ultrasound examinations, based on convolutional neural networks (CNN). To validate the effectiveness of the proposed model, experiments were conducted on 765 data samples, and four evaluation metrics-accuracy, precision, recall, and F1 score-were calculated to analyse the classification performance of the trained model. The results indicate that the new CNN model achieves a minimum recognition accuracy of no less than 70% on the test set, demonstrating robust recognition performance and meeting the designed objectives.