Light Annotation Fine Segmentation: Histology Image Segmentation Based on VGG Fusion with Global Normalisation CAM

Deep learning has been widely used to segment tumour regions in stained histopathology images. However, precise annotations are expensive and labour-consuming. To reduce the manual annotation workload, we propose a light annotation-based fine-level segmentation approach for histology images based on VGG-based Fusion network with Global Normalisation CAM. The experts only required provide rough images, then accurate boundaries can be produced using this method. validate proposed approach, three datasets fine quality built. labels as ground truth evaluation. VFGN-CAM method includes main components: an enhancement boost boundary accuracy model generalisability; VGG module that integrates multi-scale features; CAM combines local global gradient information of regions. Our fusion outperform existing methods Dice 84.188%. final improvement our against original is around 22.8%.