# Early Prediction of Hepatic Decompensation in Cirrhosis Using Optimised XGBoost Models at the Initial Outpatient Hepatology Visit

## TL;DR

This study uses machine learning to predict when cirrhosis patients might experience liver failure, helping doctors intervene earlier.

## Contribution

The paper introduces optimized XGBoost models for early prediction of hepatic decompensation using baseline clinical data.

## Key findings

- XGBoost models outperformed logistic regression in predicting decompensation at 3 and 5 years.
- Key predictors included platelet count, albumin, INR, bilirubin, age, and disease aetiology.
- Models achieved AUROC > 0.80 and high recall (0.98) at 3 and 5 years.

## Abstract

Hepatic decompensation represents a critical transition in cirrhosis, leading to increased morbidity, mortality and healthcare utilisation. Identifying patients at risk of decompensation remains a clinical challenge. We aimed to develop and validate XGBoost models to predict hepatic decompensation at multiple time points using clinical data available at a patient's initial outpatient hepatology visit.

We conducted a retrospective cohort study including 2208 adult patients with cirrhosis or its complications seen in hepatology clinics between 1985 and 2022. Patients were classified as compensated or decompensated based on a keyword search of the Philip and Ellie Kives Clinical Database, with decompensation dates confirmed by chart review. Sixteen routinely available variables including demographics, biochemical parameters and disease aetiology were used as predictors. Logistic regression and XGBoost models were trained to predict hepatic decompensation at 1, 3, 5 and 10 years, with a random 20% holdout test set used for validation. XGBoost models were tuned to optimise the precision‐recall area under the curve (PR‐AUC). Performance was evaluated using AUROC, precision, recall and F1 scores.

XGBoost models outperformed logistic regression at most time points, demonstrating strong performance at 3 and 5 years. Recall was 0.42, 0.98, 0.98 and 0.82 at 1, 3, 5 and 10 years respectively. Corresponding AUROC values were 0.85, 0.88, 0.81 and 0.89.

Optimised XGBoost models demonstrated robust predictive accuracy for medium‐ and long‐term hepatic decompensation among patients with compensated cirrhosis. These models may support early risk stratification and enable personalised management strategies to prevent clinical deterioration.

XGBoost models were developed using routinely available baseline clinical data to predict hepatic decompensation at 1, 3, 5 and 10 years in patients with compensated cirrhosis.In a cohort of 2208 patients, XGBoost models predicted hepatic decompensation with good accuracy, demonstrating AUROC > 0.80 and recall of 0.98 at 3 and 5 years.Key predictors included platelet count, albumin, INR, bilirubin, age and disease aetiology.Machine learning models optimised for Precision‐Recall AUC (PR‐AUC) may enable risk stratification that supports earlier intervention and personalised care for patients with cirrhosis.

XGBoost models were developed using routinely available baseline clinical data to predict hepatic decompensation at 1, 3, 5 and 10 years in patients with compensated cirrhosis.

In a cohort of 2208 patients, XGBoost models predicted hepatic decompensation with good accuracy, demonstrating AUROC > 0.80 and recall of 0.98 at 3 and 5 years.

Key predictors included platelet count, albumin, INR, bilirubin, age and disease aetiology.

Machine learning models optimised for Precision‐Recall AUC (PR‐AUC) may enable risk stratification that supports earlier intervention and personalised care for patients with cirrhosis.