Traditional pathology is mostly qualitative and depends on visual inspection of experts. This limitation is due to the fact that conventional light microscope allows only a view of a fraction of a specimen at a time. The selection of the view and the interpretation are thus often subject to human observer bias.

WS-recognizer is developed to better handle this problem. It is an an open-source quantitative pathology tool designed to quantify tissue characteristics across the whole slide imaging (WSI) of the histology section. WSI offers a digital replica of an entire histopathology slide, which can be process by WS-Recognizer to map, across the entire section, the tissue characteristics such as cell density and nuclear size (H&E stain), protein or antigen/antibody distribution (immunohistochemistry), axonal density and size (brain sections). It can also handle tissue microarray (TMA) slides and report number of targets for each tissue core. 

The detailed information about the method is documented in the paper, "Mapping stain distribution in pathology slides using whole slide imaging" by Yeh et al. (link).

The following is the example of the analysis result. It presents macroscopic distribution of the recognized targets, such as nuclei, RBC, ...etc. The output can be quantitatively analyzed to enable statistical analysis.