Validation of a pan-ELastography Machine-learning (ELM) score to predict clinically significant portal hypertension in compensated advanced chronic liver disease

The authors aimed to develop and validate a machine-learning model, the ELM Score, to enhance the prediction of clinically significant portal hypertension (CSPH) in patients with compensated advanced chronic liver disease (ACLD) while reducing indeterminate results. By integrating liver and spleen stiffness measurements with clinical variables, the ELM Score demonstrated superior performance compared to existing Baveno criteria, achieving high predictive values and significantly lowering the gray zone of indeterminate cases. This approach supports a safer, non-invasive identification of ACLD patients who may benefit from non-selective β-blocker therapy.