A Recurrent Neural Network and Differential Equation Based Spatiotemporal Infectious Disease Model with Application to COVID-19

A Recurrent Neural Network and Differential Equation Based Spatiotemporal Infectious Disease Model with Application to COVID-19
复制标题

基于循环神经网络和微分方程的时空传染病模型及其在 COVID-19 中的应用

DOI:
10.1101/2020.07.20.20158568
复制
发表时间:
2020
期刊:
of the 12th International Conference on Knowledge Discovery and Information Retrieval
影响因子:
--
通讯作者:
Zhou, Guofa
Zhou, Guofa
中科院分区:
--
文献类型:
--
作者:
Li, Zhijian;Zheng, Yunling;Xin, Jack;Zhou, Guofa

文献摘要

参考文献

相似文献

2019 年冠状病毒病 (COVID-19) 的爆发对世界产生了重大影响。对感染趋势进行建模并实时预测病例可以帮助决策和控制疾病传播。然而,由于每日样本的时间有限,数据驱动的方法(例如循环神经网络(RNN))可能表现不佳。在这项工作中,我们开发了一个基于流行微分方程(SIR)和 RNN 的集成时空模型。前者经过简化和离散化后是一个区域时间感染趋势的紧凑模型,而后者则模拟了最近相邻区域的影响。后者捕获潜在的空间信息。 %未公开报道。我们在意大利的 COVID-19 数据上训练和测试了我们的模型,结果表明,它在 1 天、3 天和 1 周的未来预测方面优于现有的时间模型(完全连接的 NN、SIR、ARIMA),特别是在训练数据有限的情况下。
The outbreaks of Coronavirus Disease 2019 (COVID-19) have impacted the world significantly. Modeling the trend of infection and real-time forecasting of cases can help decision making and control of the disease spread. However, data-driven methods such as recurrent neural networks (RNN) can perform poorly due to limited daily samples in time. In this work, we develop an integrated spatiotemporal model based on the epidemic differential equations (SIR) and RNN. The former after simplification and discretization is a compact model of temporal infection trend of a region while the latter models the effect of nearest neighboring regions. The latter captures latent spatial information. %that is not publicly reported. We trained and tested our model on COVID-19 data in Italy, and show that it out-performs existing temporal models (fully connected NN, SIR, ARIMA) in 1-day, 3-day, and 1-week ahead forecasting especially in the regime of limited training data.
图结构递归神经网络及其稀疏化在流行病预测中的应用研究
DOI: 10.1007/978-3-030-21803-4_73
发表时间: 2019
期刊: Algorithms and Applications
影响因子: --
作者:
LI, Zhijian;Luo, Xiyang;Wang, Bao;Bertozzi, Andrea L.;Xin, Jack
通讯作者: Xin, Jack
DOI: 10.1007/s11401-019-0168-y
发表时间: 2017-11
期刊: Chinese Annals of Mathematics, Series B
影响因子: --
作者:
Bao Wang;Penghang Yin;A. Bertozzi;P. Brantingham;S. Osher;J. Xin
通讯作者: Bao Wang;Penghang Yin;A. Bertozzi;P. Brantingham;S. Osher;J. Xin
复杂系统优化:理论、模型、算法和应用,WCGO 2019,全球优化世界大会,法国梅斯,2019 年 7 月 8-10 日
DOI: --
发表时间: 2020
期刊: World Congress on Global Optimization
影响因子: --
作者:
Hoai An Le Thi;Hoai Minh Le;T. P. Dinh
通讯作者: T. P. Dinh
DOI: 10.1038/s42005-021-00570-y
发表时间: 2021-04-20
影响因子: 5.5
作者:
Morris, Dylan H.;Rossine, Fernando W.;Levin, Simon A.
通讯作者: Levin, Simon A.