Urban Airborne and Blood Lead Analyses Using Neural Networks

Abstract

The author of this paper applies a feed-forward back-propagation neural network to airborne lead data gathered around a busy highway interchange in Birmingham, England. Data were collected during the early 1970s when lead emissions from vehicles were at their highest. The neural computation was capable of identifying realistic isopleths of lead concentrations around the interchange for summer/winter and daytime/nighttime conditions. A similar neural study on blood lead levels of residents from all over Birmingham was conducted as well. This analysis complemented the results of the original study and highlighted the importance of age, sex, location, and age of house on residents' blood levels. The author determined that linear least-squares regression analysis gave poor results and was unsuitable for interpreting this type of highly nonlinear data. The neural analyses suggested a relationship between airborne and blood lead concentrations, and indicated that opening the interchange has caused the blood lead levels in schoolboys, living adjacent to the interchange, to increase by approximately 9 percent.