Comprehensive Summary
This study explores the newfound role of geospatial artificial intelligence (GeoAI) in environmental epidemiology. GeoAI uses machine learning methods and geospatial data to create a more precise assessment of environmental exposures. Two key shifts were identified that could drive the field, first the ability to apply advanced algorithms to large geospatial data sets. This allows for a more accurate prediction of factors that could affect the environment like air and water pollution. The second shift was the widespread use of smartphones and devices that could provide location based environmental data that could be utilized. Practical applications with this data could be predicting daily exposure to fine particulate matter, indemnifying populations as risk of exposure to contaminated water sources, and using deep learning on Google Maps images to create high resolutions images of greenery and built environments. With these two shifts, researchers could get a better picture of the environmental risk for individuals and populations that goes beyond the knowledge they have now. Despite the promises, there are challenges that still play a role such as addressing privacy/consent concerns, ensuring the data is representative, and developing validation datasets to confirm the model used is accurate.
Outcomes and Implications
For clinicians, the ability to use GeoAi can provide an opportunity to get a fuller understanding of how environmental exposures contribute to health outcomes. This is especially important for diseases such as heart disease, respiratory illness, cancer, and more that effect many patients around the world. By capturing exposures at a high spatial and temporal resolution, GeoAI can play a role in the epidemiological research and clinical decision making. This aids in early detecting, risk management, and personalized prevention that makes care more precise for each patient. However, it's important for the medical community to also be aware of the challenge that comes with the utilization of GeoAI. Incomplete or biased datasets could skew the effects or treatment of a demographic group of patients leading to incorrect health conclusions. Ethical guidelines could also play a role in determining how GeoAI can be applied responsibly and in compliance with ethics guidelines across different disciples. Moreover, integrating GeoAI into health research and setting can help shift the field to be more personalized and prevention based, especially after addressing the drawback in the system as it stands.