Back

A quantitative comparison between human experts and AI at estimating tumor-stroma ratio

Pai, A.; Brunekreef, J.; Marcus, E.; Sanders, J.; Guimaraes, M. D. S.; Menezes, R.; Sanchez, C. I.; Horlings, H. M.; Teuwen, J.

2025-08-28 pathology
10.1101/2025.08.26.25334451 medRxiv
Show abstract

The tumor-stroma ratio (TSR) is an established prognostic biomarker across several cancer types, yet its manual assessment remains labour-intensive and subject to inter-observer variability. An artificial intelligence (AI)-based estimate could offer an efficient, consistent alternative for this task. In this study, quantitative comparisons were made between expert humans and an AI model for TSR estimation. Using two independent, multi-institutional histopathology datasets, an Attention U-Net was benchmarked against experienced pathologists. In a subset of the TCGA-BRCA dataset, the AI model demonstrated comparable trends to human consensus for TSR quantification, achieving an intraclass correlation coefficient (ICC) of 0.69. However, the AI models TSR scores are on average 5 percentage points higher compared to human scores on this dataset. The AI model was found to be more consistent at estimating TSR than either of the human counterparts, with a discrepancy ratio (DR) of 0.86. Results on an external dataset obtained from the Netherlands Cancer Institute consisting of cases (n=357) from 35 different Dutch hospitals showed that the AI models TSR scores are 7 percentage points lower on average compared to the human rater with an ICC of 0.59. To account for the models imperfect segmentation performance, we derived an estimate of the ambiguity in AI-based TSR predictions. The results indicate that, despite this ambiguity, the AI not only follows similar trends but also delivers greater overall scoring consistency than manual TSR assessment by humans.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.