Skip to main content
. Author manuscript; available in PMC: 2021 Dec 2.
Published in final edited form as: Phys Med Biol. 2021 Aug 27;66(17):10.1088/1361-6560/ac176d. doi: 10.1088/1361-6560/ac176d

Figure 1.

Figure 1.

Illustration of SDL-Seg structure. The model consists of three main steps: 1. Saliency map generation where saliency maps with target location cues are produced, 2. Saliency-guided U-Net encoding the regions with high saliency for more accurate segmentation of breast target. 3.Majority voting fuses predictions of multiple U-Nets to generate final segmentation