Leveraging AI and transfer learning to enhance out-of-hospital cardiac arrest outcome prediction in diverse setting.

Access to trustworthy artificial intelligence (AI) for clinical applications is uneven, especially in low-resource settings with limited and inconsistent data. Models from high-resource settings often fail to generalize. Transfer learning (TL) can adapt established models to new settings. Using neurological outcome prediction for out-of-hospital cardiac arrest (OHCA) as a proof of concept, we adapted a model trained on a large cohort to Vietnam (243 patients) and Singapore (15,916 patients) using the Pan-Asian Resuscitation Outcomes Study registry. The external model performed poorly on the Vietnam cohort, with an area under the receiver operating characteristic curve (AUROC) of 0.467 (95% CI: 0.141-0.785), but TL markedly improved performance (AUROC = 0.807, 95% CI: 0.626-0.948). In Singapore, TL yielded modest gains (AUROC = 0.955 vs. 0.945). These findings highlights the potential of TL to improve prediction accuracy across diverse healthcare contexts and to support equitable and safe global AI adoption.