Tumor–Stroma Ratio in Colorectal Cancer—Comparison between Human Estimation and Automated Assessment

SIMPLE SUMMARY: A lower tumor–stroma ratio within a tumor correlates with a poorer outcome, i.e., with a higher risk of death. The assessment of this ratio by humans is prone to errors, and when presented the same case, the ratios reported by multiple pathologists will oftentimes deviate significantly. The aim of our work was to predict the tumor–stroma ratio automatically using deep neural segmentation networks. The assessment comprises two steps: recognizing the different tissue types and estimating their ratio. We compared both steps individually to human observers and showed that (i) the outlined automatic method yields good segmentation results and (ii) that human estimations are consistently higher than the automated estimation and deviate significantly for a hand-annotated ground truth. We showed that including an additional evaluation step for our segmentation results and relating the segmentation quality to deviations in tumor–stroma assessment provides helpful insights. ABSTRACT: The tumor–stroma ratio (TSR) has been repeatedly shown to be a prognostic factor for survival prediction of different cancer types. However, an objective and reliable determination of the tumor–stroma ratio remains challenging. We present an easily adaptable deep learning model for accurately segmenting tumor regions in hematoxylin and eosin (H&E)-stained whole slide images (WSIs) of colon cancer patients into five distinct classes (tumor, stroma, necrosis, mucus, and background). The tumor–stroma ratio can be determined in the presence of necrotic or mucinous areas. We employ a few-shot model, eventually aiming for the easy adaptability of our approach to related segmentation tasks or other primaries, and compare the results to a well-established state-of-the art approach (U-Net). Both models achieve similar results with an overall accuracy of 86.5% and 86.7%, respectively, indicating that the adaptability does not lead to a significant decrease in accuracy. Moreover, we comprehensively compare with TSR estimates of human observers and examine in detail discrepancies and inter-rater reliability. Adding a second survey for segmentation quality on top of a first survey for TSR estimation, we found that TSR estimations of human observers are not as reliable a ground truth as previously thought.