Extraction table for a systematic review of AI/ML models for early prediction of infectious disease outbreaks (2024–2025)

Study-level extraction table underlying a systematic review of artificial-intelligence and machine-learning models for the early prediction of infectious disease outbreaks using climatic, sociodemographic and open data. The dataset contains one row per included study (n = 41 peer-reviewed studies published in 2024–2025) and 22 coded fields covering bibliographic details, target disease, data sources, model family and techniques, best reported performance metric, the explainability (XAI) coding matrix (approach, technique and scope), the PROBAST risk-of-bias appraisal by domain and overall, and citation counts. Coding convention: absence of reporting is coded as absence of the property, not as missing data. A study was coded as providing an explanation only where the technique, its target and its scope were explicitly stated in the text, tables or supplementary material. Reported performance metrics are not comparable across studies and citation counts are snapshots, not quality measures. Full definitions of every field are provided in the data_dictionary sheet of the XLSX file. The review was registered in PROSPERO (CRD420261420067, retrospectively registered).