Coarse-to-Fine Stacked Fully Convolutional Nets for lymph node segmentation in ultrasound images

摘要: Ultrasound as a well-established imaging modality is widely used in lymph nodes for clinical diagnosis and disease analysis. Quantitative analysis of node features, morphology, relations can provide valuable information immune system studies. For such analysis, it necessary to first accurately segment the areas ultrasound images. In this paper, we develop new deep learning method, called Coarse-to-Fine Stacked Fully Convolutional Nets (CFS-FCN), automatically segmenting Our method consists multiple stages FCN modules. We train CFS-FCN model learn segmentation knowledge from coarse-to-fine, simple-to-complex manner. A data set 80 images containing both normal diseased our experiments, which show that considerably outperforms state-of-the-art methods segmentation.