Skip to main content
. 2021 Aug 2;13(15):3891. doi: 10.3390/cancers13153891

Figure 2.

Figure 2

The proposed deep learning framework for segmentation of PTC from papanicolaou-stained thyroid FNA and TP WSIs. (a) The workflow of the proposed deep learning framework. Initially, each WSI is formatted into hierarchical data structure and processed by fast background filtering and the proposed deep learning model is used to produce the segmentation results of PTC on papanicolaou-stained thyroid FNA and TP WSIs. (b) The detailed architecture of the proposed deep learning model.