Cross-Map of AI and Infectious Disease Research: A Panoramic Analysis of the Scoping-Review Database

The Scoping-Review project has built a searchable evidence map database for the application of artificial intelligence in infectious diseases and clinical microbiology. Using the scoping review method, it systematically organizes relevant literature and provides researchers with a multi-dimensional annotated literature retrieval tool to facilitate interdisciplinary communication and the development of medical AI research.

Infectious disease prevention and control is an eternal challenge in public health. AI technology has brought new tools to this field, but interdisciplinary research is currently scattered. A scoping review can depict the overall landscape of the field, identifying hotspots and gaps; evidence maps, through structured annotation and visualization, meet the needs of researchers from different backgrounds (computer scientists, clinicians, policymakers) and help them quickly establish domain knowledge.

The core of the database is a multi-dimensional non-exclusive tagging system. Each record includes tags such as publication year, AI methods (machine learning/deep learning, etc.), application types (diagnosis/prediction, etc.), disease types (COVID-19/tuberculosis, etc.), and clinical/laboratory focus. The non-exclusive design allows a single literature to have multiple tags, truly reflecting the characteristics of interdisciplinary research and supporting fine-grained filtering.

• Researcher navigation: Multi-dimensional filtering to quickly locate relevant literature (e.g., deep learning + tuberculosis + diagnosis);
• Interdisciplinary bridge: Structured annotations provide a common language framework;
• Policy reference: Analyzing literature distribution to guide resource allocation (e.g., identifying under-researched areas).

The database supports keyword/tag combination searches and may use relational or document databases. It requires efficient indexing and a good user interface (multi-condition combination, sorting, export). Data needs to be continuously updated, and a literature monitoring mechanism should be established to ensure timeliness.

• Coverage boundary (e.g., may focus on English literature);
• Annotation subjectivity requires quality control (multi-person annotation, consistency check);
• In the future, it can be connected to open-access literature databases to provide full-text links.

The project is open-source and transparent. The screening and annotation process can be viewed, and secondary analysis is supported. Community members can report errors, supplement literature, and suggest new dimensions. Crowdsourced maintenance makes up for resource shortages.

The Scoping-Review database is a knowledge infrastructure for AI infectious disease research. It lowers the entry barrier to the field, promotes communication, and guides research directions. We look forward to its continuous update and expansion to contribute to public health, and its methodology is worth learning from for other interdisciplinary fields.