Air pollution by chemical or bacteriological pollutants is not limited to a geographical area at the discrete border. Also, factors taken into account in the risk analysis such as patient age, seasons, socio-economic background or race are far from being necessary as discrete values. Often used analysis tools are limited to statistical analyzes. Given the nature of this data, imprecise and complex, in this study we propose an artificial intelligence tool, in particular the principles of fuzzy logic in data processing. A fuzzy system is constructed with five fuzzy input variables (Nature of the pollutant, geographic area, season,) and an output variable which expresses the corresponding health risk in terms of disease with its incidence rate. The rule base must contain all possible combinations. This tool can be used as a tool to aid in the prognosis and in the prevention of the onset of epidemiological diseases.