Artificial intelligence for predicting chronicity, relapse, and zoonotic risk in brucellosis: toward a one health closed-loop decision-making framework
Abstract
Brucellosis prediction spans persistent clinical disease, post-treatment recurrence, and transmission across human-animal-environment interfaces. This narrative review distinguishes primary prediction studies from supporting risk-factor evidence and adds a report-based appraisal using PROBAST for individual prognosis and adapted domains for nonclinical models. The evidence is uneven: one reviewed chronicity model showed moderate internal discrimination; relapse evidence chiefly concerns label quality and treatment-related risk factors; and forecasting studies predict reported cases rather than individual infection. Genomic models classify proxy zoonotic labels and require independent confirmation. Different outcome definitions, limited validation, and ascertainment bias prevent pooled estimates or direct performance ranking. We propose a staged One Health framework that begins with routine clinical and surveillance data and adds animal, environmental, or genomic inputs only when accessible and demonstrably useful. The full loop remains a proposal rather than a validated intervention. Priorities are reproducible endpoints, prospective external validation, calibration, feasible cross-sectoral governance, and measured clinical or public-health benefit. Affordability and cost-effectiveness remain hypotheses to be tested.
Related
- Predicting follow-up cognition using estimated structural and functional connectivity networks and artificial intelligence in multiple sclerosis
- Predicting Pathological Dental Wear: A Roadmap for Artificial Intelligence in Natural Teeth and Restorations
- Artificial Intelligence in Delirium Care: Predicting, Detecting, and Preventing Delirium in Older Hospitalized Adults.
- Artificial intelligence-enhanced electrocardiography for predicting mortality and major adverse cardiovascular events: A systematic review and narrative synthesis
- Application of Artificial Intelligence for Predicting Sports Injuries and Customizing Personalized Prevention Strategies: A Scoping Review
BibTeX
@article{wang2026artificial,
title = {{Artificial intelligence for predicting chronicity, relapse, and zoonotic risk in brucellosis: toward a one health closed-loop decision-making framework}},
author = {Juanxia Wang and Zhiquan Cheng},
journal = {Frontiers in Public Health},
year = {2026},
month = nov,
doi = {10.3389/fpubh.2026.1907782},
url = {https://www.frontiersin.org/articles/10.3389/fpubh.2026.1907782},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Artificial intelligence for predicting chronicity, relapse, and zoonotic risk in brucellosis: toward a one health closed-loop decision-making framework
AU - Juanxia Wang
AU - Zhiquan Cheng
PY - 2026
DA - 2026/11/03
JO - Frontiers in Public Health
DO - 10.3389/fpubh.2026.1907782
UR - https://www.frontiersin.org/articles/10.3389/fpubh.2026.1907782
AB - Brucellosis prediction spans persistent clinical disease, post-treatment recurrence, and transmission across human-animal-environment interfaces. This narrative review distinguishes primary prediction studies from supporting risk-factor evidence and adds a report-based appraisal using PROBAST for individual prognosis and adapted domains for nonclinical models. The evidence is uneven: one reviewed chronicity model showed moderate internal discrimination; relapse evidence chiefly concerns label quality and treatment-related risk factors; and forecasting studies predict reported cases rather than individual infection. Genomic models classify proxy zoonotic labels and require independent confirmation. Different outcome definitions, limited validation, and ascertainment bias prevent pooled estimates or direct performance ranking. We propose a staged One Health framework that begins with routine clinical and surveillance data and adds animal, environmental, or genomic inputs only when accessible and demonstrably useful. The full loop remains a proposal rather than a validated intervention. Priorities are reproducible endpoints, prospective external validation, calibration, feasible cross-sectoral governance, and measured clinical or public-health benefit. Affordability and cost-effectiveness remain hypotheses to be tested.
SN - 2296-2565
ER -