The incidence rate and mortality of lung cancer continuously rise, and lung cancer is one of the leading causes of cancer death deaths. Detecting the high prevalence areas of lung cancer can effectively provide clue for detecting its influential factors. Taking the incidence rate of lung cancer in Xiamen as an example, this paper adopted Moran's I and Getis's G statistics to detect global correlation characteristics and hot spot areas of the incidence rate of lung cancer based on street (town) scale and district scale. And the incidence rate of lung cancer in different gender and ages were analysed in hot spot areas. The result shows that spatial correlation characteristics of the incidence rate of lung cancer based on district scale are not obvious, and spatial correlation characteristics of the incidence rate of lung cancer based on street (town) scale are obvious, and the hot spot areas of the incidence rate of lung cancer are detected. Another result shows that the incidence rate of lung cancer in different gender and ages is different in different hot spot areas. The conclusion is that it is very important to select a proper spatial scale for the research of the incidence rate of lung cancer. And the main age groups of lung cancer are 50~59 years old, 60~69 years old and 70~79 years old, and lung cancer incidence of men is higher than women.