Colorectal Tumor Identification by Transferring Knowledge from Pan-cytokeratin to HE

摘要

Tumor budding is a recently recognized, independent prognostic factor in colorectal cancer, but lacks astandardized assessment methodology. Although staining with pan-cytokeratin has been shown to mitigatethe issue of lack of reproducible, intra-observer agreement, usage of this antibody remains expensive andis limited in clinical practice. We propose an automated image analysis framework to take advantage of thevisual superiority of pan-cytokeratin and the routine use of H&E to detect and quantify tumor budding. Ourframework has demonstrated promising ability to identify tumor regions of colorectal slides – 92.0%accuracy, 94.5% sensitivity, and 85% specificity – across four independent datasets.