Digital twins based on bidirectional LSTM and GAN for modelling the COVID-19 pandemic

Quilodrán-Casas, César; Silva, Vinicius Santos; Arcucci, Rossella; Heaney, Claire E.; Guo, Yike; Pain, Christopher C.

doi:10.48550/arxiv.2102.02664

Cited by 2 publications

(8 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The test case used here is the spatio-temporal variation of a virus infection in an idealized town. The extended SEIRS model [29,26] is a nonlinear model with compartments of susceptible (S), exposed (E), infectious (I), and recovered (R). It extends the traditional theory of the dynamics of infectious diseases [3,9,8] to account for variations not only in time but also in space.…”

Section: Test Case Descriptionmentioning

confidence: 99%

“…The initial condition is that 0.1% of people at home have been exposed to the virus and will thus develop an infection. The epidemiological parameters used in this work are chosen to be consistent with those of COVID-19, and are described in [29,26]. Further information about the discretization and solution methods of the high-fidelity numerical simulation can be also found in [29,26].…”

Section: Problem Set Upmentioning

confidence: 99%

“…In this context, surrogate models are computationally appealing and have been attracting significant attention in the last decades. Deep neural networks have become one of the most popular surrogate models in science and engineering nowadays [39,35,42,40,33,26]. Among them, generative adversarial networks (GAN) have been demonstrating promising results.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

GAN for time series prediction, data assimilation and uncertainty quantification

Silva¹,

Heaney²,

Pain³

2021

Preprint

Self Cite

View full text Add to dashboard Cite

We propose a new method in which a generative adversarial network (GAN) is used to quantify the uncertainty of forward simulations in the presence of observed data. Previously, a method has been developed which enables GANs to make time series predictions and data assimilation by training a GAN with unconditional simulations of a high-fidelity numerical model. After training, the GAN can be used to predict the evolution of the spatial distribution of the simulation states and observed data is assimilated. In this paper, we describe the process required in order to quantify uncertainty, during which no additional simulations of the high-fidelity numerical model are required. These methods take advantage of the adjoint-like capabilities of generative models and the ability to simulate forwards and backwards in time. Set within a reduced-order model framework for efficiency, we apply these methods to a compartmental model in epidemiology to predict the spread of COVID-19 in an idealised town. The results show that the proposed method can efficiently quantify uncertainty in the presence of measurements using only unconditional simulations of the high-fidelity numerical model.Preprint. Under review.

show abstract

Section: Test Case Descriptionmentioning

confidence: 99%

Section: Problem Set Upmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

GAN for time series prediction, data assimilation and uncertainty quantification

Silva¹,

Heaney²,

Pain³

2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Where possible, parameters of the model were chosen to be consistent with those of COVID-19. With compartments of susceptible (S), exposed (E), infectious (I) and recovered (R), the extended SEIRS equations [25] are used to generate the highfidelity numerical simulations that describe the spread of infections in both space and time. Based on differential equations, multi-compartment SEIR-type models [26,27] can be very costly to solve: there may be millions of variables every time step; and the time steps may need to be small to model the movement of people around the domain.…”

Section: Introductionmentioning

confidence: 99%

“…In order to reduce the computational cost of numerical simulations, ROMs are now commonly used for applications in computational physics [28,29]. Nonetheless, they are relatively new to virus modelling [25]. A ROM is a lowdimensional representation of a high-dimensional model or discretised system, and should be accurate enough for the desired use and at least several orders of magnitude faster to solve than the high-dimensional system.…”

Section: Introductionmentioning

confidence: 99%

Data Assimilation Predictive GAN (DA-PredGAN): applied to determine the spread of COVID-19

Silva,

Heaney,

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

We propose the novel use of a generative adversarial network (GAN) (i) to make predictions in time (PredGAN) and (ii) to assimilate measurements (DA-PredGAN). In the latter case, we take advantage of the natural adjoint-like properties of generative models and the ability to simulate forwards and backwards in time. GANs have received much attention recently, after achieving excellent results for their generation of realistic-looking images. We wish to explore how this property translates to new applications in computational modelling and to exploit the adjoint-like properties for efficient data assimilation. To predict the spread of COVID-19 in an idealised town, we apply these methods to a compartmental model in epidemiology that is able to model space and time variations. To do this, the GAN is set within a reduced-order model (ROM), which uses a low-dimensional space for the spatial distribution of the simulation states. Then the GAN learns the evolution of the low-dimensional states over time. The results show that the proposed methods can accurately predict the evolution of the high-fidelity numerical simulation, and can efficiently assimilate observed data and determine the corresponding model parameters.

show abstract

Digital twins based on bidirectional LSTM and GAN for modelling the COVID-19 pandemic

Cited by 2 publications

References 0 publications

GAN for time series prediction, data assimilation and uncertainty quantification

GAN for time series prediction, data assimilation and uncertainty quantification

Data Assimilation Predictive GAN (DA-PredGAN): applied to determine the spread of COVID-19

Contact Info

Product

Resources

About