Skip to main content
Fig. 5 | Acta Neuropathologica Communications

Fig. 5

From: Toward a generalizable machine learning workflow for neurodegenerative disease staging with focus on neurofibrillary tangles

Fig. 5

Inference results for NFT detection. a Example WSI from the Emory-Holdout cohort with Pre-NFT (blue) and iNFT (red) predictions. Predictions are mostly localized to the gray matter regions of the tissue (outer edge), as is expected for NFTs since neurons are mostly present in these regions. b At high resolution we can see the distinct differences between the two classes of predictions, with iNFTs being more fibrillary and darker in color while Pre-NFTs being putative in texture, showing a clear nuclei, and a lighter brown. NFT Detection on WSIs

Back to article page