A Transfer Learning-Based Multivariate Control Chart for Dengue Surveillance in Hong Kong
Dengue is a severe mosquito-borne epidemic disease. There is no effective vaccine for dengue, so a real-time surveillance system becomes crucial to detect dengue outbreaks. Control charts have been widely used as efficient tools to identify changes in health-related data. In Hong Kong, the environme...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2023-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10167630/ |
| Summary: | Dengue is a severe mosquito-borne epidemic disease. There is no effective vaccine for dengue, so a real-time surveillance system becomes crucial to detect dengue outbreaks. Control charts have been widely used as efficient tools to identify changes in health-related data. In Hong Kong, the environmental protection department uses the area ovitrap index to survey monthly the number of mosquitoes in different areas. Some areas have limited historic area ovitrap index records since the survey started only recently. Parameter estimation for designing control charts is challenging with little historic data. This paper proposes a transfer learning-based estimator to increase parameter estimation accuracy for areas with limited historic data points. We study the questions on what and how to transfer useful knowledge from related source data. A multivariate control chart based on transfer-learned parameters is developed for online monitoring. A real example of dengue surveillance demonstrates its effectiveness in application. |
|---|---|
| ISSN: | 2169-3536 |