Development and validation of a machine learning scoring system based on multi-dimensional metabolic signatures for invasive examination triage in intestinal diseases.
- Open access
A machine learning scoring system using metabolic biomarkers achieves an AUC of 0.910 for IBD diagnosis, surpassing traditional fecal calprotectin tests.
- Why it matters: Accurate, non-invasive tools are needed to distinguish IBD from IBS, reducing unnecessary colonoscopies and optimizing clinical resource use amid overlapping symptoms.
- What they did: Researchers analyzed 729 participants, collecting clinical and metabolism-related biomarkers, and applied correlation analysis, LASSO regression, and XGBoost to develop and validate a scoring system.
- The result: The scoring system reliably differentiates IBD, correlates with inflammation severity, and offers a practical, non-invasive approach to triage invasive examinations, enhancing diagnostic efficiency.