Gut decisions based on the liver: prediction of colorectal neoplasia using AI-based liver analysis of routine CT scans

1

This study explores AI-based liver analysis from routine CT scans as a non-invasive method for predicting colorectal neoplasia.
2

Data from 1,997 patients were analyzed, revealing a significant predictive capability of liver imaging features for colorectal neoplasia.
3

The XGBoost model achieved an AUROC of 0.810, significantly outperforming a clinical-only model with an AUROC of 0.457.
4

Sensitivity and specificity for detecting colorectal neoplasia were 74.1% and 72.3%, respectively, after threshold optimization.
5

The findings support the gut-liver axis as a novel biomarker source, enhancing colorectal cancer screening strategies.

Frontiers In Oncology

by Anna Hinterberger, Jonas Bohn, Darya Trofimova, Nicolas Knabe, Julia Dettling, Tobias Norajitra, Fabian Isensee, Johannes Betge, Stefan O. Schönberg, Dominik Nörenberg, Sergio Grosu, Sonja Loges, Ralf Floca, Jakob Nikolas Kather, Klaus Maier-Hein, Freba Grawe
June 3, 2026

1