Disease surveillance by clustering based on minimal internal distance

Disease surveillance is essential for studying disease spread. An important task in disease surveillance is identifying disease clusters, which are areas of unusually high incidence. In this paper, we formulate the disease surveillance problem as a clustering problem and review some standard techniques used for clustering problems. Inspired by techniques used in graph theory, we introduce our new method, which is based on a new statistic derived from minimal internal distance in the graph, to solve this problem. Simulated and real lung cancer data from New Mexico are analyzed according to our method, and results are compared with those of the popular spatial scan statistic.