Machine learning techniques called deep learning, specifically convolutional neural networks, have revolutionized medical image analysis. Before deep learning, machine learning relied on extracting features from images, but now deep learning can directly learn from image data. This approach has significantly improved the performance of medical image analysis, allowing for better classification of lesions, segmentation of organs, and detection of abnormalities. Deep learning is expected to become the dominant technology in medical image analysis in the coming decades, offering higher accuracy and more advanced capabilities.