Abstract
Rabies is an important but often neglected zoonotic disease with significant public health, veterinary, and economic impacts. Inadequate surveillance and diagnostic resources in low- and middle-income countries impede its effective public health responses. This study demonstrates the use of an extreme gradient boosting (XGB) technique for animal rabies risk assessment in endemic areas of Haiti with limited diagnostic resources. We employed spatiotemporal clustering and trend analysis techniques to assess rabies status across large geographic areas. The XGB model achieved high specificity (0.99, 95% CI: 0.98–0.99), sensitivity (0.78, 95% CI: 0.61–0.95), negative predictive value (1.0, 95% CI: 0.99–1.00), and accuracy (0.98, 95% CI: 0.98–0.99) in predicting animal rabies cases. The framework identified 20 high-risk rabies clusters, representing a 40% increase in detected transmission zones compared to laboratory-confirmed data alone. This included 8 clusters in underserved communities where no diagnostic infrastructure was previously available, enabling real-time monitoring of disease and surveillance trends in previously unmonitored regions. Trend analysis facilitated real-time monitoring of high-risk clusters with significantly changing disease and surveillance statuses. This study showcases a data-driven, evidence-based framework for neglected and emerging zoonotic disease surveillance in economically constrained regions.
Similar content being viewed by others
Machine learning to improve the understanding of rabies epidemiology in low surveillance settings
Higher free-roaming dog density sustains rabies virus transmission in Haiti
Enhancing rabies epidemic modeling with neural networks and fractional calculus
Acknowledgements
The authors wish to thank the late Mr. Christian Morange who played an integral role in establishing the Haiti Animal Rabies Surveillance Program, investigating nearly 2,000 suspected rabid animals, nearly 20% of the data used in this study.
Funding
The Integrated Bite Case Management program in Haiti is supported by funding from the Global Health Security Agenda.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Competing interests
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Ethical approval
All methods were carried out in accordance with relevant guidelines and regulations. This study involving human participants was reviewed and approved by the CDC as not to be research, as the purpose of these activities was to investigate and assess a condition of public health importance. CDC Institutional Review Board (IRB) review was not required under the following provision: Public health surveillance activities, 45 CFR 46.102(l)(2). Written informed consent for participation was not required for this study in accordance with the Ministère de la Santé Publique et de la Population (MSPP) in Haiti and the IRB at the U.S. CDC (Public health surveillance activities, 45 CFR 46.102(l)(2)). The animal study was reviewed and approved by CDC Institutional Animal Care and Use Committee (IACUC). Written informed consent for participation was not obtained from the owners because the purpose of this surveillance program is to monitor the occurrence of animal rabies cases in Haiti through the investigation of animal bites and through the testing of pathology samples from deceased animals. The surveillance program is a component of the national rabies control program, which includes vaccination programs, PEP guidelines, and animal quarantine and euthanasia policies. Surveillance will be used to inform ongoing practice within the national rabies control program. As an activity designed to monitor the occurrence of disease in a defined population, as well as to provide feedback to inform ongoing public health practice, this activity is consistent with the attributes of non-research public health practice, as described in current CDC policy.
Additional information
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Below is the link to the electronic supplementary material.
Supplementary Material 1 (download PDF )
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Reprints and permissions
About this article
Cite this article
Keshavamurthy, R., Blanton, J., Orciari, L. et al. Biosurveillance and early outbreak detection of rabies in settings with limited laboratory capacity using spatiotemporal clustering and a machine learning framework.
Sci Rep (2026). https://doi.org/10.1038/s41598-026-55346-7
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-026-55346-7
Keywords
- Rabies
- Zoonoses
- Machine learning
- Disease surveillance
- Extreme gradient boosting
- Spatiotemporal clustering
- Situational awareness
Source: Ecology - nature.com
