Lonny Chen Rong Qi Ahmad1, Sarbhan Singh1, Balvinder Singh Gill1, Jane Labadin2*, Boon Hao Hong2, Wei King Tiong2, Cia Vei Tan1, Sumarni Mohd Ghazali1, Jenarun Jelip3, Norhayati Mokhtar3
1. Biomedical Epidemiology Unit, Special Resource Centre, Institute for Medical Research, National Institutes of Health, Setia Alam, 40170, Selangor, Malaysia
2. Department of Computational Science and Mathematics, University of Malaysia Sarawak, 94300, Kota Samarahan, Sarawak, Malaysia
3. Disease Control Division, Ministry of Health Malaysia, 62590 Putrajaya, Malaysia
*Corresponding author: Jane Labadin.
ABSTRACT
INTRODUCTION: While dengue control typically targets residential areas through fogging, COVID-19 lockdowns have demonstrated a 30% case reduction, indicating the importance of non-residential dengue transmission. Existing systems lack the spatial precision to identify the source of dengue infection.
OBJECTIVE(S): To develop MozzHub, a web application that identifies dengue infection sources at a 400-meter resolution using patient mobility and environmental data through bipartite network modeling.
MATERIALS & METHODS: The system analyzes patients' 14-day pre-symptom movement history, incorporating environmental factors (temperature, humidity, precipitation) and mosquito survival parameters to rank potential infection locations. User acceptance testing involved 31 participants from six Malaysian district health offices using a validated 16-item questionnaire based on the Technology Acceptance Model.
RESULTS: MozzHub successfully identified potential dengue sources of infection, requiring only mobility data input while performing all preprocessing and analysis automatically. The system achieved high user acceptance scores: 80.6% for ease of use, 77.4% for usefulness, and 74.2% for user satisfaction. Functional testing confirmed proper operation of heatmap visualization, case entry, and movement record processing.
CONCLUSION: MozzHub represents a meaningful advancement in dengue surveillance methodology by integrating automated micro-scale spatial analysis with human mobility data. The demonstrated high user acceptance and operational efficiency support its viability for deployment in resource-constrained settings. This system has the potential to significantly improve vector control effectiveness by enabling targeted interventions at identified transmission sources rather than conventional residential-focused approaches.
KEYWORDS: Dengue Surveillance, Bipartite Network, Hotspot Detection, Spatial Analysis, Public Health Informatics


