Weakly supervised learning for multi-organ adenocarcinoma classification in whole slide images

Abstract

AbstractThe primary screening by automated computational pathology algorithms of the presence or absence of adenocarcinoma in biopsy specimens (e.g., endoscopic biopsy, transbronchial lung biopsy, and needle biopsy) of possible primary organs (e.g., stomach, colon, lung, and breast) and radical lymph node dissection specimen is very useful and should be a powerful tool to assist surgical pathologists in routine histopathological diagnostic workflow. In this paper, we trained multi-organ deep learning models to classify adenocarcinoma in biopsy and radical lymph node dissection specimens whole slide images (WSIs). We evaluated the models on seven independent test sets (stomach, colon, lung, breast, lymph nodes) to demonstrate the feasibility in multiorgan and lymph nodes specimens from different medical institutions and international public datasets, achieving receiver operating characteristic areas under the curves (ROC-AUCs) in the range of 0.91-0.99.