Fully Automated Tumor Bud Assessment in Hematoxylin and Eosin Stained Whole Slide Images of Colorectal Cancer.

Tumor budding (TB), the presence of single cells or small clusters of up to four tumor cells, at the invasive front of colorectal cancer (CRC) is a proven risk factor for adverse outcomes. International definitions are necessary to reduce the interobserver variability. According to the current international guideline, hotspots at the invasive front should be counted in Hematoxylin and Eosin (H&E) stained slides. This is time-consuming and prone to interobserver variability, therefore there is a need for computer-aided diagnosis solutions. In this paper, we report on developing an Artificial Intelligence (AI) based method for detecting tumor budding in H&E-stained whole slide images. We propose a fully automated pipeline to identify the tumor border, detect tumor buds, characterize them based on their number of tumor cells, and produce a TB density map that we use to identify the TB hot spot. The method outputs the TB count in the hotspot as a computational biomarker. We show that the proposed automated TB detection workflow performs on par with a panel of five pathologists at detecting tumor buds, and that the hotspot-based TB count is an independent prognosticator in both the univariate and the multivariate analysis, validated on a cohort of n=981 CRC patients. Computer-aided detection of tumor buds based on deep learning can perform on par with expert pathologists at detection and quantification of tumor buds in H&E-stained colorectal cancer histopathology slides, strongly facilitating the introduction of budding as an independent prognosticator in clinical routine and clinical trials.