Table 1 AI predictive model proposals
From: Should AI allocate livers for transplant? Public attitudes and ethical considerations
AI model prediction | Example proposals | General relevant findings |
|---|---|---|
Pre-transplant mortality (urgency) | Cucchetti, 2007; Bertsimas, 2019 | |
Post-transplant survival | Cruz-Ramírez, 2013; Briceño, 2014; Lau, 2017; Matis, 1995; Ayllón, 2018; Ershoff, 2020; Haydon, 2005; Hoot, 2005; Khosravi, 2015; Dorado-Moreno, 2017; Zhang, 2012 | Predict survival at 30 days, 3 months, 1–5 years post-transplant [2, 28,29,30,31,32,33,34,35,36,37]. Most more accurate than current scoring systems. |
Survival benefit (e.g., life years gained) | Schaubel, 2009; Dancs, 2022 | Predict survival benefit (using pre-and post-transplant survival predictions) [38, 39]. |
QALYs | Santos, 2020 | Predict 30-day quality-adjusted survival (in critically ill cancer patients) [40]. None in organ allocation: feasible area of future research. |