tetano
Editor, Senior Moderator
Emerg Infect Dis
. 2020;26(11):2733-2735.
doi: 10.3201/eid2611.200706.
Thresholds versus Anomaly Detection for Surveillance of Pneumonia and Influenza Mortality
Timothy L Wiemken, Ana Santos Rutschman, Samson L Niemotka, Daniel Hoft
Abstract
Computational surveillance of pneumonia and influenza mortality in the United States using FluView uses epidemic thresholds to identify high mortality rates but is limited by statistical issues such as seasonality and autocorrelation. We used time series anomaly detection to improve recognition of high mortality rates. Results suggest that anomaly detection can complement mortality reporting.
Keywords: data science; immunization; infection; influenza; machine learning; pneumonia; respiratory infections; seasonality; surveillance; time series; vaccine; vaccine-preventable diseases; viruses.
. 2020;26(11):2733-2735.
doi: 10.3201/eid2611.200706.
Thresholds versus Anomaly Detection for Surveillance of Pneumonia and Influenza Mortality
Timothy L Wiemken, Ana Santos Rutschman, Samson L Niemotka, Daniel Hoft
- PMID: 33079038
- DOI: 10.3201/eid2611.200706
Abstract
Computational surveillance of pneumonia and influenza mortality in the United States using FluView uses epidemic thresholds to identify high mortality rates but is limited by statistical issues such as seasonality and autocorrelation. We used time series anomaly detection to improve recognition of high mortality rates. Results suggest that anomaly detection can complement mortality reporting.
Keywords: data science; immunization; infection; influenza; machine learning; pneumonia; respiratory infections; seasonality; surveillance; time series; vaccine; vaccine-preventable diseases; viruses.