Three-Month Real-Time Dengue Forecast Models: An Early Warning System for Outbreak Alerts and Policy Decision Support in Singapore
Problem
Singapore, with endemic dengue and no vaccine on the market, needed forecasts of epidemic evolution up to three months ahead to support public health response.
Why it matters
The 2013 outbreak reached 22,170 cases, and early warning enables timely vector control and policy decisions.
What the AI did
Developed LASSO-based statistical models forecasting weekly dengue notifications over a 3-month horizon using case, meteorological, vector surveillance and population data, updated weekly.
Results
not stated in source
Impact
The abstract describes a weekly-updated forecasting tool for outbreak alerts and policy support; accuracy figures are not stated.
Record
| Venue | Environmental Health Perspectives |
|---|---|
| Published | 2015 |
| Work type | article |
| Authors | Yuan Shi, Xu Liu, Suet Yheng Kok, Jayanthi Rajarethinam, Shaohong Liang, Grace Yap, Chee‐Seng Chong, Kim-Sung Lee et al. |
| Citations | 140 (OpenAlex, at import time) |
| References | 46 |
| Model or method | LASSO regression models |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1289/ehp.1509981 |
| Open access | gold |
| Keywords | dengue forecasting, early warning system, LASSO, Singapore, vector control |
Importance rating
significant — thousands to millions affected
Endemic dengue with no vaccine; 2013 outbreak reached 22,170 cases; early warning supports vector control and public health policy.