Wavelet-Based Monitoring for Biosurveillance

Abstract

Biosurveillance, focused on the early detection of disease outbreaks, relies on classical statistical control charts for detecting disease outbreaks. However, such methods are not always suitable in this context. Assumptions of normality, independence and stationarity are typically violated in syndromic data. Furthermore, outbreak signatures are typically of unknown patterns and, therefore, call for general detectors. We propose wavelet-based methods, which make less assumptions and are suitable for detecting abnormalities of unknown form. Wavelets have been widely used for data denoising and compression, but little work has been published on using them for monitoring. We discuss monitoring-based issues and illustrate them using data on military clinic visits in the USA

TQWT-assisted statistical process control method for condition monitoring and fault diagnosis of bearings in high-speed rail

“…The Shewhart control chart with control limits computed from each scale can be used to monitor the process, and this method is also called multiscale statistical process control (MSSPC). 14 When the wavelet coefficients on different scales are independent of each other, then a statistic combining all the decomposed coefficients can be constructed for monitoring. For example, in the research work of Jung and Kim, 15 a statistic was constructed by combining all the decomposed wavelet coefficients after discrete wavelet transform (DWT) to monitor the electrocardiographic signal to diagnose the premature ventricular contraction beats.…”