The Spatiotemporal Surveillance Method Based On Least Angle Regression

Public health surveillance involves collecting data on chronic or infectious diseases and monitoring changes in disease incidences in order to improve public health.Diseases data are collected on time and contain regional information,and it is critical to detect outbreaks and find locations of disease in time in public health.Most of the existing studies focus on the detection of onset times or areas and models are too idealistic to be applied in practice.An Exponential Weighted Moving Average(EWMA)control chart based on Least Angle Regression is proposed in this paper to monitor complex forms of outbreak and areas simultaneously.The simulation results based on male thyroid cancer data in New Mexico show that the control chart presented in this paper has a good performance in detecting the time of outbreak and judging the areas.