Related Experiment Video
Updated: Jun 14, 2025
10:56
Long-term Behavioral Tracking of Freely Swimming Weakly Electric Fish
Published on: March 6, 2014
12.5K
Poisson Kalman filter for disease surveillance
Donald Ebeigbe1, Tyrus Berry2, Steven J Schiff1,3
1Center for Neural Engineering, Department of Engineering Science and Mechanics, Pennsylvania State University, University Park, Pennsylvania 16802, USA.
Physical Review Research
|August 30, 2024
Summary
A new optimal filter, based on the Kalman filter, is developed for analyzing infectious disease data. This method enhances the tracking of disease dynamics, including epidemics like COVID-19.
Area of Science:
- Epidemiology
- Biostatistics
- Control Theory
Background:
- Poisson distributions are commonly used to model daily infectious disease case counts.
- Traditional Kalman filters may not optimally handle the characteristics of Poisson-distributed observational data in disease surveillance.
Purpose of the Study:
- To develop an optimal filter for Poisson observations, extending the Kalman filter framework.
- To apply the developed filter to real-world disease data, including neonatal sepsis and hydrocephalus.
Main Methods:
- Development of a linear and a nonlinear (extended) optimal filter for Poisson data.
- Application of the filter to a case study using parameters from publicly available data.
Main Results:
- The optimal filter provides an effective method for analyzing infectious disease dynamics.
- The approach demonstrated applicability to both noncommunicable and communicable diseases.
Conclusions:
- The proposed filter variant offers an improved approach for modeling infectious disease surveillance data.
- This method is adaptable for a wide spectrum of diseases, including epidemics like COVID-19.
