MATHEMATICAL MODEL OF DENGUE TRANSMISSION INVOLVES MOBILITY AND CLIMATE FACTORS USING CLUSTERING METHOD

• Main Author: Fakhruddin, Muhammad
• Format: Dissertations
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/51324
• Institution: Institut Teknologi Bandung

Dengue is a disease that can be fatal and has been frequently reported to increase dramatically in recent years. Based on reports from the World Health Organization (WHO), dengue is also included in the list of ten diseases that cause the highest mortality in the world. The emergence of outbreaks every year in many countries indicates the complexity of dengue prevention and control. It is a big challenge for researchers and policymakers to understand the spread and establish dengue early warning systems. Climatology factors such as rainfall, humidity, and temperature are known as the primary determinant in the intensity of the dengue incidence. The appropriate conditions of rainfall, humidity, and temperature can help vectors to multiply optimally in their ecosystem. This climate information then plays a critical role in the spread of the dengue virus as the main factor supporting vector dynamics, which is difficult to predict. On the other hand, increasing population density and human mobility allow viruses to spread and multiply in non-endemic areas. This study discusses mathematical models of dengue virus transmission that involve mobility and climate factors using a clustering method. First, host-vector deterministic and stochastic models in closed populations are reviewed and validated using dengue incidence data. However, the autonomous models can only cover the data in an outbreak period. Therefore, a non-autonomous dengue model was constructed to cover and analyze the seasonal effects of dengue incidences and their infection rates. This periodicity is also shown by dengue incidence and climate data that have been reported for several years by Health Office and Meteorology, Climatology and Geophysics Agency (BMKG), respectively. Climate factors are then involved in constructing of the dengue transmission model using multiple regression models integrated with a clustering method. The climate variables used in this study are rainfall and humidity. This clustering is used to distinguish the effects of different climatic conditions on dengue cases. The optimal barriers are then obtained by using a Particle Swarm Optimization method, which minimizes the residual. Furthermore, the dengue dynamics model involving climate factors contained in the infection rate parameter and other parameters in vector dynamics are constructed to analyze the relationship between host-vector dynamics and climate. At the end of this study, a dengue dynamics model was constructed by involving the mobility and climate factors integrated with a clustering method, which can be used as dengue prediction. This study can be used as an early effort to prevent dengue spatially and temporally. The spatial distribution pattern and the spatial correlation of dengue incidences were investigated at the end of this study. This study is very beneficial for the relevant health offices and policymakers as consideration for making optimal and efficient intervention efforts.