Dai Tran Le Vuong, Viet Vu Duc, Bao Minh Phuong Long, Hung Pham Xuan, Dat Tran Vuong Quoc, Thanh Nguyen Ha, Linh Nguyen Khanh, Khoa Nguyen Dinh, Cuong Do Duy, Duc Nguyen Dinh

Main Article Content

Abstract

Artificial Intelligence (AI) is emerging as a strategic solution to healthcare challenges in developing countries, such as Vietnam, where systems face overcrowding, workforce shortages, and quality demands. This scoping review aims to explore the current applications and research directions of AI in Vietnamese healthcare. Following the Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) guidelines, we searched international databases (PubMed, Google Scholar, and ScienceDirect) and major Vietnamese medical journals, identifying 37 relevant studies from 2018 to 2024. They were categorized into six groups: diagnostic imaging support, complication and treatment prediction, community screening, healthcare chatbots, smart health record management, and drug development assistance. These findings show that Vietnam has made initial progress in applying AI, especially in diagnostic imaging. However, most of the current models focus on analyzing isolated images that lack data integration. Consequently, these models are limited to disease detection without identifying the underlying causes. To address the existing gaps, we propose the development of a multimodal AI model for pneumonia etiology diagnosis by integrating chest X-rays with clinical and laboratory data. In conclusion, our findings highlight Vietnam’s initial progress and propose future directions for AI in precision medicine, contributing to antibiotic resistance control and healthcare resource optimization.