TITLE:
A Novel ICT-Enabled Decision Support Approach for Surveillance and Control of Mosquito-Borne Diseases
AUTHORS:
Aman Hassan Bura, Emmanuel Milambo Mung’onya
KEYWORDS:
Mosquito-Borne Diseases, Digital Health Surveillance, HL7 FHIR, GraphQL Middleware, PostGIS, ACSSI, ABAC
JOURNAL NAME:
E-Health Telecommunication Systems and Networks,
Vol.15 No.1,
March
24,
2026
ABSTRACT: Mosquito-borne diseases continue to impose a substantial public health burden in low- and middle-income countries, where surveillance is hindered by fragmented data systems, limited laboratory capacity, and unreliable network connectivity. Conventional REST-based digital health platforms often require multiple endpoint calls and redundant data transfers, resulting in increased latency, inefficient bandwidth utilization, and delayed epidemiological response. This paper presents an interoperable ICT-enabled surveillance architecture that integrates clinical diagnostics, geospatial intelligence, and automated mosquito identification through a GraphQL-mediated middleware and a low-cost Automated Computer-Supported Specimen Imaging (ACSSI) edge node. The ACSSI node performs on-site specimen imaging and lightweight AI-based classification using commodity off-the-shelf (COTS) and predominantly open-source hardware components customized for local deployment, thereby reducing reliance on centralized laboratory microscopy. Experimental evaluation demonstrates consistent improvements over REST, including latency reductions of 35% - 40%, throughput increases of 20% - 30%, and backend CPU utilization reductions of 10% - 15%. The embedded classifier achieved 92% accuracy, 90% precision, 88% recall, and an F1-score of 0.89, confirming reliable field performance. Although automation introduces moderate initial deployment costs, overall cost-effectiveness is achieved through lower operational expenditure enabled by COTS hardware, minimal maintenance requirements, reduced bandwidth consumption, and decreased dependence on specialized clinical personnel. These results demonstrate a scalable, sustainable, and operationally cost-efficient framework for real-time mosquito-borne disease surveillance in resource-constrained settings.