TL/DR –
The FibroX, an explainable AI model, has demonstrated superior performance in early detection of advanced fibrosis and predicting all-cause and cardiovascular mortality in patients with metabolic dysfunction-associated steatotic liver disease (MASLD). The FibroX model was tested against the fibrosis-4 index (FIB-4) in a study presented at Digestive Disease Week 2025, and showed improved accuracy, interpretability, and potential cost savings, including the prevention of 16.5 million unnecessary VCTEs and $3.3 billion in US healthcare costs. The most common cause of chronic liver disease, MASLD, is expected to become the leading indication for liver transplant in the US, with predictions of a steady increase in prevalence from 33.7% in 2020 to 41.4% by 2050.
FibroX AI Model Improves Advanced Fibrosis Detection in MASLD Patients
Recent research underlines the value of FibroX, an explainable AI model, in enhancing advanced fibrosis detection and predicting overall and cardiovascular mortality in patients suffering from metabolic dysfunction-associated steatotic liver disease (MASLD).
The research, presented at the Digestive Disease Week (DDW) 2025 by Basile Njei, MD, MPH, PhD, from Yale School of Medicine, showcases the accuracy, interpretability, and cost-effectiveness of the novel AI model in comparison to the fibrosis-4 index (FIB-4).
Njei and colleagues highlight the limited accuracy of current noninvasive liver disease assessment tools like FIB-4, used to identify MASLD patients at risk for advanced fibrosis. MASLD, the most common cause of chronic liver disease, is predicted to become the leading reason for liver transplant in the US.
FibroX: A Solution to Current Shortcomings
FibroX was developed to overcome these limitations by improving advanced fibrosis detection and predicting overall and cardiovascular mortality. Its utility was tested in a cohort of adults with MASLD from NHANES 2017–2020, following 2024 AASLD guidelines.
Using eXtreme Gradient Boosting (XGBoost), the FibroX model was internally validated with five-fold cross-validation and a 95% specificity cutoff to reduce false positives. Clinical performance was evaluated by reviewing the charts of 100 MASLD patients with liver biopsy results. This was followed by external validation using an NHANES 1988–1994 cohort to predict mortality over 30 years.
Key predictors of cardiovascular mortality were identified using Shapley Additive Explanations (SHAP), and a cost analysis was conducted to assess the economic impact of FibroX as a first-line screening tool in the US.
Superior Performance of FibroX
Results demonstrated that FibroX outperformed FIB-4 in detecting advanced fibrosis, with superior accuracy in biopsy-confirmed cases. Additionally, FibroX predicted cardiovascular mortality, unlike FIB-4. Cost analysis revealed that FibroX could prevent 16.5 million unnecessary VCTEs, resulting in $3.3 billion savings in US healthcare costs.
The researchers concluded that FibroX is highly accurate and beneficial in detecting advanced liver fibrosis in MASLD, enhancing long-term cardiovascular mortality predictions, thereby underscoring its potential for integration into clinical workflows to improve patient management.
References
- Njei B, Ilagan-Ying YC, Boateng S, et al. FIBROX: AN EXPLAINABLE AI MODEL FOR ACCURATE PREDICTION OF ADVANCED LIVER FIBROSIS AND CARDIOVASCULAR MORTALITY IN MASLD. Digestive Disease Week 2025.
- Le P, Tatar M, Dasarathy S, et al. Estimated Burden of Metabolic Dysfunction–Associated Steatotic Liver Disease in US Adults, 2020 to 2050. JAMA Netw Open. doi:10.1001/jamanetworkopen.2024.54707
—
Read More Health & Wellness News ; US News