A machine learning-based model to predict multi-time-point prognosis for acute-on-chronic hepatitis B liver failure
Jieyang Jin1,2, Zhong Liu3, Mei Liao1,2
1Department of Ultrasound, the Third Affiliated Hospital of Sun Yat-Sen University, 600 Tianhe Road, Guangzhou, P.R. China.
Iscience
|April 9, 2026
Summary
This study introduces a prognostic model and mobile app for evaluating acute-on-chronic hepatitis B liver failure. It aids in patient care and liver transplant decisions.
Area of Science:
- Hepatology
- Clinical Medicine
- Medical Informatics
Background:
- Acute-on-chronic hepatitis B liver failure (ACHBLF) presents complex prognostic challenges.
- Accurate prognostication is crucial for timely clinical decision-making, especially regarding liver transplantation.
Purpose of the Study:
- To develop and validate a multi-time-point prognostic model for ACHBLF.
- To create a user-friendly mobile application for implementing the prognostic model in clinical practice.
Main Methods:
- Development of a prognostic scoring system based on clinical and laboratory parameters.
- Validation of the model using retrospective patient data.
- Integration of the model into a mobile application for accessibility.
Main Results:
- The prognostic model demonstrated significant accuracy in predicting patient outcomes at multiple time points.
- The mobile application provided a practical tool for real-time prognostic evaluation.
- Improved clinical decision-making regarding liver transplantation was observed.
Conclusions:
- The prognostic model and mobile application offer a valuable tool for managing patients with ACHBLF.
- This integrated approach enhances patient care and supports informed decisions for liver transplantation.
- Further prospective validation is warranted to confirm long-term clinical utility.
