Summary of Forecasting Infectious Disease Prevalence with Associated Uncertainty Using Neural Networks, by Michael Morris
Forecasting infectious disease prevalence with associated uncertainty using neural networks
by Michael Morris
First submitted to arxiv on: 2 Sep 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Populations and Evolution (q-bio.PE)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary Medium Difficulty Summary: This thesis proposes two methodological frameworks for accurately forecasting infectious disease incidence, particularly influenza-like illness (ILI) in the United States. The first framework combines Web search activity data and historical ILI rates as observations for training neural networks (NNs), incorporating Bayesian layers to produce uncertainty intervals. The iterative recurrent neural network (IRNN) model outperforms the state-of-the-art by reducing mean absolute error by 10.3% and improving Skill by 17.1%. The second framework uses neural ordinary differential equations to bridge the gap between mechanistic compartmental models and NNs, utilizing a mixture of ILI rates and Web search activity data for forecasting. Eight neural ODE models are evaluated, with those trained without Web search activity data outperforming the IRNN0 by 16% in terms of Skill. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Low Difficulty Summary: This research helps predict when diseases will spread and how many people will get sick. The goal is to make it easier for public health agencies to respond quickly to new or emerging diseases. Two new methods are proposed using special kinds of computer models called neural networks (NNs). One method uses information from the internet, like what people search online, to help predict when diseases will spread. Another method combines different types of data, like how many people got sick in the past and what people searched online, to make more accurate predictions. The best approach was found to be a model that reduced errors by 10% and improved accuracy by 17%. This research can help save lives and reduce the economic burden of diseases. |
Keywords
» Artificial intelligence » Neural network
