A review on recent advancements of ChatGPT and datafication in healthcare applications

Integrating ChatGPT, a large language model (LLM), with datafication strategies presents the capacity to provide reliable and user-friendly healthcare applications. This paper provides a structured and critical review of how ChatGPT, supported by datafication and big data frameworks, shapes areas such as clinical decision support, telemedicine, and personalized healthcare. It examines the cognitive capabilities of ChatGPT in interpreting complex medical language while addressing key ethical concerns, such as algorithmic bias, data privacy, and explainability, that arise in its deployment. Moreover, the study highlights challenges related to data quality, provenance, and interoperability, which are vital for real-world implementation. The review synthesizes recent literature, identifies research gaps, and explores policy and technical considerations necessary for responsible adoption. Rather than asserting promising outcomes, the paper contextualizes the capabilities and limitations of ChatGPT in current healthcare settings, aiming to inform future developments in AI-driven healthcare systems by bridging theoretical perspectives with early-stage implementations. The paper aims to guide future developments in AI-driven healthcare systems by bridging theoretical perspectives with emerging use cases.