Analysis of air pollution (PM10) and respiratory morbidity rate using K-maximum sub-array (2-D) algorithm

Abstract

Air pollution is known to cause adverse effects on human health. While pidemiological studies generally involve statistical/geographical approaches, the use of computer algorithms is yet uncommon. In this study, the K-Maximum Sub-array (K-MSA) algorithm is introduced for the first time in health and environmental science to analyse the association between particulate air pollution with diameter less than 10 μm (PM10), from four years of daily measurements (1998-2002), and acute respiratory hospital admission counts over a wide range of age groups (0 to 98 years) in Christchurch, New Zealand. The morbidity rate was taken from residents within a 2 km diameter of the air pollution monitoring site, located in a residential area. The K-MSA algorithm identifies K maximum threshold values, e.g., what age groups associate most with specific ranges of [PM10] levels, by maximizing the sum of the elements of a selected sub-array of a two-dimensional array. The K-MSA successfully detects that the associations with different [PM10] levels vary between different age groups, and that responses even differ between sexes, and different time period (winter, summer or all seasons). The K-MSA will be an encouraging methodology to investigate how various air pollution levels are related to health or climate in the future to increase knowledge for future air pollution policy making. Copyright 2007 ACM.