Smart Community Health Monitoring and Early Warning System for Water-Borne Diseases in Rural

View at publisher Summarise with VIRAT
Authors: P. Preethi, Sri Vishruthi C, Revathi R
Cited by: 0
References: 9

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Water-borne diseases continue to be a major concern in rural regions, primarily due to inadequate monitoring of water quality and delayed identification of outbreaks. Recognizing the recurring nature of such diseases emphasizes the importance of proactive health surveillance in these communities. This study introduces an Intelligent Community Health Monitoring and Early Warning Framework that integrates Spatio-Temporal Graph Neural Networks (ST-GNN) with advanced Boosting algorithms, including XGBoost and LightGBM. By analyzing both patient health records and environmental factors, the proposed model aims to predict potential outbreaks, thereby connecting data- driven research with practical public health applications. By integrating rural health surveillance, anomaly detection, and patient data prediction, the proposed framework can generate early warnings, allowing healthcare professionals to intervene before diseases become widespread. This proactive approach addresses a critical gap in low-resource settings. Ultimately, the aim of this system is to improve community health monitoring, reduce response times, and support health authorities in making data-driven decisions in resource-limited regions. From an academic perspective, it demonstrates how data-science techniques can be effectively applied to enhance the quality of life in rural communities. In addition to conventional monitoring approaches, the proposed framework integrates a dynamic AI-driven pipeline capable of adapting to evolving disease patterns over time. The Spatio-Temporal Graph Neural Network (ST-GNN) component captures intricate spatial relationships among rural wards, while ensemble models such as XGBoost and LightGBM strengthen the temporal analysis, enabling accurate outbreak predictions. Incorporating real-time patient and environmental data enhances the system's responsiveness and interpretability. Experimentally, the model achieved an F1-score of 0.71 to 0.79 and a sensitivity of 0.69, demonstrating its effectiveness for real-world health surveillance in rural settings. This framework exemplifies how AI-based hybrid systems can transition from experimental research to practical early-warning applications in resource-limited regions

Fields of Study & Subjects

Data-Driven Disease Surveillance Public Health and Environmental Issues Artificial Intelligence in Healthcare

Publication details

Journal / source
International Conference on Communication, Computing and Electronics Systems (ICCCES)
Pages
909–914
Publication date
21 January 2026
Language
EN