Generative AI as a Catalyst for Interoperability and Data-Driven Decision Support in Healthcare Systems of Developing Countries

Interoperability across heterogeneous information systems remains a persistent challenge, particularly in resource constrained contexts where infrastructures are fragmented and data formats remain incompatible. This study introduces a novel methodology that integrates generative Artificial Intelligence (AI) with the urbanization of information systems to enable scalable and seamless interoperability. The approach employs AutoGen AI, an open-source orchestration framework powered by Large Language Models (LLMs), specifically GPT-4o, to coordinate task-specific intelligent agents for data extraction, transformation, and harmonization. By converting disparate data into a standardized JSON representation, the architecture resolves both syntactic and semantic inconsistencies while simultaneously emphasizing multi-agent concurrency, distributed orchestration, and computational scalability, resulting in improved through-put, reduced latency, and enhanced robustness. A real-world healthcare case study is presented to illustrate the framework’s effectiveness: heterogeneous clinical datasets were unified into a coherent JSON structure, enabling accurate health indicator generation and reliable decision support. Experimental results demonstrate substantial improvements in system connectivity, processing efficiency, and integration reliability, with potential to generalize far beyond the medical sector. Moreover, the methodology incorporates advanced prompt engineering and context-aware dialogue design, minimizing model hallucinations and ensuring trustworthy outputs in LLM-driven processes. Overall, the study positions generative AI not only as a promising solution for interoperability in health informatics, but also as a transformative paradigm for intelligent system integration across diverse domains characterized by distributed, heterogeneous environments.