[2I5-OS-7a] OS-7

3:30 PM - 3:45 PM JST(6:30 AM - 6:45 AM UTC)

[2I5-OS-7a-01] Development of a Prognostic Prediction System for Acute Liver Failure/Liver Injury: Risk Scoring

〇Taiyo Oura¹, Raiki Yoshimura¹, Shingo Iwami¹ (1. Nagoya University)

Keywords:

Clinical Score,Risk Assessment,Disease Progression

Acute liver failure (ALF) and acute liver injury (ALI) are rapidly progressive and potentially fatal conditions, often requiring liver transplantation. However, given the limited availability of donor organs, there is an urgent need for clinically applicable and reliable prognostic tools to identify patients who truly require transplantation. In this study, we developed a machine learning–based AI model to predict the need for liver transplantation using admission blood test data from 319 patients with various causes of ALI. Furthermore, we constructed a statistically grounded risk scoring system with comparable predictive performance, optimized for practical use in clinical settings. This score enables early prediction of future disease severity at the time of admission and supports timely treatment decisions. Implementation of this system could improve patient outcomes by facilitating the efficient allocation of limited medical resources. Our team is also concurrently developing a clinical decision support app based on this risk score.