Research Article

TSHVNet: Simultaneous Nuclear Instance Segmentation and Classification in Histopathological Images Based on Multiattention Mechanisms

Figure 1

TSHVNet structure diagram. (a) A residual unit, (b) a dense unit, (c) a SimAM module, and (d) a Transformer attention module. The size of the original image is , and the center point is for identification. The size of the output image is also . The output image and the center patch of the surrounding part are spliced to form a image.