Environmental change, especially rapid urbanization, has measurable effects on human health. Over 70 percent of the global population is expected to live in urban areas by 2025, and at least 1 in 3 of these individuals will live in extreme poverty. Living in a slum or shantytown exposes individuals to health and safety risks including inadequate sanitation, lack of access to clean water, air pollution, violence, over-crowding and risk of infectious disease. Remote sensing technologies are an effective tool for detection and prediction of areas with significant land use change, which enables identification of populations with the highest risk of various adverse health outcomes. This project aims to contribute toward the development of a remote sensing-based, integrated infectious disease prediction and surveillance program for Puerto Maldonado, Peru, a city with expected rapid urbanization resulting from the newly constructed Peru-Brazil Interoceanic highway. The first stage of this study evaluated landscape change in Puerto Maldonado between 2002 and 2011. Results indicate that significant changes occurred during this time, including considerable loss of vegetation and the expansion of built structures and roadways. These changes occurred simultaneously with the construction of a major trade route through the city. The second stage of this study investigated the accuracy of pixel-based techniques versus object-based techniques to classify land use and land cover attributes, such as pasture-land and human dwellings, of a high-resolution, multispectral WorldView2 image. The results of this study show that a pixel-based approach has a higher degree of accuracy (93.6%) than an object-based approach (89.4%) on high-resolution 8-band imagery. This suggests that with higher-resolution imagery and increased spectral data, incorporating segment information in an object-based analysis may not be necessary and may actually decrease classification accuracy. Public health researchers have only recently begun to use remote sensing techniques in their efforts to combat vector-borne disease. Elucidation of the simplest and most efficient classification methods is a necessary step toward using these models in the fight against vector-borne disease. By identifying ground conditions that expose individuals to a higher risk of a specific disease, targeted and more cost-effective prevention and control programs can be implemented.