Integrating Natural Language Processing and Large Language Models Into DHIS2 to Improve Health Data Utilization

In the evolving landscape of global healthcare, the need for efficient and scalable data analysis has never been more critical. This research explores the integration of Large Language Models (LLMs) and Natural Language Processing (NLP) into the District Health Information System 2 (DHIS2), a platform widely used for health data management in Uganda. The objective of this study is to enhance the usability and data analysis capabilities of DHIS2, making it accessible to both technical and non-technical users. By leveraging AI-driven models like LangChain and Ollama, the research developed an automated data analysis system that allows users to interact with DHIS2 using natural language queries in English. This integration simplifies the extraction of meaningful insights from complex datasets, thereby improving decision-making processes. The system was evaluated for usability, performance, and its potential to enhance public health data analysis. Results demonstrate that the AI-augmented DHIS2 platform significantly reduces analysis time, improves data accessibility, and facilitates timely public health interventions. This innovation presents a scalable solution for improving health information systems globally, particularly in resource-constrained environments.