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.

Authors

Institutions

Publication Details

Journal
npj Digital Medicine
Published
2025-11-21
DOI
https://doi.org/10.1038/s41746-025-02088-x
Primary Topic
Cardiac Arrest and Resuscitation
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

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

Wei Ling Tay, Dai Quoc Khuong, Desmond Renhao Mao, Hung Quoc To et al.
npj Digital Medicine
Cardiac Arrest and Resuscitation
article

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

Wei Ling Tay, Dai Quoc Khuong, Desmond Renhao Mao, Hung Quoc To, Ling Tiah, Lai Peng Tham, Hai Bang Truong, Quang Duy Pham, Thang X. Vu, Quoc TA Hoang, Andrew Fu Wah Ho, Michael Yih Chong Chia, Shir Lynn Lim, Wei Ming Ng, Long Hoang Le, Do Ngoc Son, Yih Yng Ng, Huan Huu Nguyen, Yohei Okada, Si Oon Cheah, Michael Hao Chen, Chinh Quoc Luong, Marcus Eng Hock Ong, Siqi Li, Nan Liu, Dat Nguyen, Tuan Anh Nguyen, Wenjun Gu, Trang Thuy Nguyen, Nausheen Edwin Doctor, Shun Yee Low, Benjamin Sieu‐Hon Leong, Hung Trong Nguyen, Han Nee Gan
article en

Abstract

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.

npj Digital MedicineVol. 8(1)
SingHealth (SG), National University of Singapore (SG), Duke University (US), Kyoto University (JP), Singapore General Hospital (SG), Changi General Hospital (SG), Tan Tock Seng Hospital (SG), Khoo Teck Puat Hospital (SG), Ng Teng Fong General Hospital (SG), Vinh Phuc Department of Health (VN), Hue Central Hospital (VN), National University Heart Centre Singapore (SG), KK Women's and Children's Hospital (SG), Care International (CH), Duke-NUS Medical School (SG), Hung Vuong Hospital (VN), National University Hospital (SG), Bạch Mai Hospital (VN), Viet Duc Hospital (VN), Children's Hospital 1 (VN), Sengkang General Hospital (SG)
American University of Beirut, ZOLL Medical Corporation, Duke-NUS Medical School, Chonnam National University, Kyungpook National University, Seoul National University, Laerdal Foundation for Acute Medicine, National Taiwan University Hospital, National Taiwan University, National Cheng Kung University, Peking University, Kansai Medical University, National Taiwan University Hospital Yunlin Branch, College of Medicine, Seoul National University, Medical Research Council, National Medical Research Council, Japan Society for the Promotion of Science
Openalex Percentile: Top 49%
Cardiac Arrest and Resuscitation
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.