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

作者：Siqi Li, Yohei Okada, Wenjun Gu, Michael Hao Chen, Do Ngoc Son, Quyet Dinh Pham, Quoc TA Hoang, Marcus Eng Hock Ong, Nan Liu, for the PAROS Investigators, Michael Yih Chong Chia, Yih Yng Ng, Benjamin Sieu‐Hon Leong, Han Nee Gan, Desmond Renhao Mao, Wei Ming Ng, Nausheen Edwin Doctor, Ling Tiah, Andrew Fu Wah Ho, Wei Ling Tay, Si Oon Cheah, Shun Yee Low, Lai Peng Tham, Shir Lynn Lim, Dai Quoc Khuong, Long Hoang Le, Tuan Anh Nguyen, Chinh Quoc Luong, Thang X. Vu, Dat Nguyen, Huan Huu Nguyen, Hung Quoc To, Hai Minh Truong, Hung Trong Nguyen, Trang Thuy Nguyen · 发表于：npj Digital Medicine · 年份：2025 · DOI：10.1038/s41746-025-02088-x · 被引用次数：2 · 研究领域：Cardiac Arrest and Resuscitation、Sepsis Diagnosis and Treatment、Artificial Intelligence in Healthcare and Education

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.