Introduction:
Artificial Intelligence (AI) has achieved remarkable performance in medical image analysis, particularly in tasks such as object detection, segmentation, and classification. In this paper, we introduce a solution for automatic breast cancer diagnosis based on the U-Net architecture, which we call (U-Net)+. The novel (U-Net)+ is designed to handle both segmentation and classification tasks within a single framework. We retained the original U-Net architecture due to its strong learning capabilities and advantages in semantic segmentation. Notably, we incorporated fully connected layers into the bottleneck layers, serving as a multi-functional classifier for both initial diagnoses based on raw images and further diagnoses for segmented images. The (U-Net)+ model is trained using a joint loss function. We conducted experiments on breast ultrasound images, demonstrating that the (U-Net)+ performs well in both classification and segmentation tasks.
Achievements:
The paper, From U-Net to (U-Net) +, what innovations have we made for the treatment and discovery of breast cancer? was published in the 13th International Conference on Artificial Intelligence, Soft Computing and Applications (AAIA), 2023 (listed in Publications).
The author Miinuo presented the work successfully at AAIA 2023.
Media