Smart Community Health Monitoring System for Water-Borne Diseases
Mr. M. Mohanasundharam1, Suvin Rajan J.Y2, Yogesh S3, Sivaa Ananthan V.A4, Sivakaruppanandh P.S5
1Assistant Professor, Department of Computer Science & Engineering, Hindusthan College of Engineering & Technology, Coimbatore
Email: mohanasundharam.cse@hicet.ac.in
2UG Scholar, Department of CSE, Hindusthan College of Engineering & Technology, Coimbatore
Email: 720723104157@hicet.ac.in
3UG Scholar, Department of CSE, Hindusthan College of Engineering & Technology, Coimbatore
Email: 720723104188@hicet.ac.in
4UG Scholar, Department of CSE, Hindusthan College of Engineering & Technology, Coimbatore
Email: 720723104146@hicet.ac.in
5UG Scholar, Department of CSE, Hindusthan College of Engineering & Technology, Coimbatore
Email: 720723104147@hicet.ac.in
Abstract — Water-borne diseases continue to pose a severe public health threat in densely populated communities across South Asia, accounting for an estimated 485,000 diarrheal deaths annually worldwide. This paper proposes and evaluates a Smart Community Health Monitoring System (SCHMS) that integrates Internet of Things (IoT) sensor networks, machine learning (ML) classification algorithms, and real-time cloud-based analytics to enable proactive detection and containment of water-borne disease outbreaks. The system continuously measures physicochemical water quality parameters—including turbidity, pH, total dissolved solids (TDS), dissolved oxygen, nitrate concentration, and coliform bacterial load—at distributed community water points. A Random Forest ensemble classifier, trained on a dataset of 42,000 annotated water quality samples from 18 Indian municipalities, achieves a disease-risk prediction accuracy of 94.7% with a sensitivity of 96.2% and specificity of 93.1%. The proposed SCHMS incorporates an SMS/app-based community alert module, a geospatial risk-mapping dashboard for health authorities, and an automated water-treatment feedback loop. Field trials conducted across six peri-urban wards of Coimbatore Municipal Corporation over a 12-month period demonstrated a 38% reduction in reported water-borne illness cases and a mean alert-to-response latency of 4.3 minutes. The system provides a scalable, low-cost architecture suitable for rapid deployment in resource- constrained municipal settings.
Keywords — Water-borne diseases; IoT sensor network; Machine learning; Community health surveillance; Water quality monitoring; Disease outbreak detection; Smart public health; Random Forest classifier.
