Fig. 1From: Pathologist-trained machine learning classifiers developed to quantitate celiac disease features differentiate endoscopic biopsies according to modified marsh score and dietary intervention responseA and E Normal duodenal tissue and biopsy sample with a modified Marsh score of 3b (H&E), respectively; B and F) same samples showing the tissue classifier overlay (villous epithelium in pink, crypt epithelium in purple, lamina propria in green and submucosa with Brunners glands in red, white space in blue); C) and G) CD3 IHC; D and H) slightly higher magnification showing the CD3 IHC overlay image with nuclei in blue and positive cells with yellow outlinesBack to article page