Pemanfaatan Artificial Intelligence Dalam Deteksi Dini, Early Warning System, dan Sistem Rujukan Kegawatdaruratan Pada Resource-Limited Settings: A Narrative Review

Background: Healthcare facilities with limited resources face various challenges in early detection and emergency referral systems, including shortages of healthcare personnel, inadequate infrastructure, and the limited accuracy of conventional screening methods. Artificial Intelligence (AI) has the potential to serve as a solution to improve the quality of clinical decision-making under these conditions. Objective: This scientific paper aims to identify and synthesize current evidence regarding the use of AI in early detection, early warning systems, and triage. Methods: This study is a narrative review using the PubMed and Google Scholar databases. English-language articles published between 2019 and 2026 that discussed the implementation of AI in early detection or emergency referral systems in resource-limited settings were included. Of the 2,376 articles identified, 16 met the inclusion criteria. Results: The articles were predominantly focused on early detection and early warning systems (50%), followed by triage and referral systems (31%), and health system strengthening (19%). Machine Learning and Deep Learning were the most commonly used AI methods. Several studies demonstrated high performance, with AUC values ranging from 0.80 to 0.97. AI was shown to improve the accuracy of sepsis detection, patient prioritization, and support faster and more objective referral decision-making. Conclusion: AI has the potential to enhance early detection, triage, and emergency referral systems in healthcare facilities with limited resources, although its implementation continues to face challenges related to infrastructure, user trust, as well as regulatory and data governance aspects.