Application of artificial intelligence in healthcare sector: A case study of predictive analytics model development for humanitarian healthcare system

Access to quality healthcare is a persistent challenge for refugee populations globally, particularly in conflict-affected regions with limited resources. The use of Artificial Intelligence (AI) might ease the challenge. Thus, this article explores the development of a predictive analytics model for forecasting patient outcomes within the healthcare system operated by the United Nations Relief and Works Agency for Palestine Refugees in the Near East (UNRWA) in Syria. A Machine Learning (ML) model development process was used to build a predictive analytics model capable of identifying patients at high risk of hospital readmission within 30 days of discharge, using synthetic fromdata with 1000 patients. The model was evaluated and achieved an accuracy of 88%, suggesting its potential effectiveness in supporting early intervention and resource prioritization in humanitarian health settings. Thus, the model can be a potential tool for data-driven decision-making that can improve healthcare outcomes within the UNRWA health system. To provide insights for scholars and practitioners, a framework titled ‘Humanitarian ML Readiness Framework (H-MLRF)’ is proposed. The framework suggests key activities and their deliverables for developing and applying predictive analytics models in humanitarian settings. Thus, the article contributes to discourse on development of predictive analytics and ML models.