COVID-19 Prevention Strategies for Victoria Students within Educational Facilities: An AI-Based Modelling Study

Abstract

Educational institutions play a significant role in the community spread of SARS-CoV-2 in Victoria. Despite a series of social restrictions and preventive measures in educational institutions implemented by the Victorian Government, confirmed cases among people under 20 years of age accounted for more than a quarter of the total infections in the state. In this study, we investigated the risk factors associated with COVID-19 infection within Victoria educational institutions using an incremental deep learning recurrent neural network-gated recurrent unit (RNN-GRU) model. The RNN-GRU model simulation was built based on three risk dimensions: (1) school-related risk factors, (2) student-related community risk factors, and (3) general population risk factors. Our data analysis showed that COVID-19 infection cases among people aged 10-19 years were higher than those aged 0-9 years in the Victorian region in 2020-2022. Within the three dimensions, a significant association was identified between school-initiated contact tracing (0.6110), vaccination policy for students and teachers (0.6100), testing policy (0.6109), and face covering (0.6071) and prevention of COVID-19 infection in educational settings. Furthermore, the study showed that different risk factors have varying degrees of effectiveness in preventing COVID-19 infection for the 0-9 and 10-19 age groups, such as state travel control (0.2743 vs. 0.3390), international travel control (0.2757 vs. 0.3357) and school closure (0.2738 vs. 0.3323), etc. More preventive support is suggested for the younger generation, especially for the 10-19 age group.

Keywords: COVID-19; artificial intelligence; deep learning; educational facilities; epidemiology; infection control; neural networks.

Figure 1

Risk factors classification for preventing COVID-19 infection among Victoria student within education facilities. The school-associated risk factors for COVID-19 infection encompass the various factors and policies in the school environment or initiated by the school that influence the infection among students. The student-related community risk factors refer to the factors and policies that affect the COVID-19 infection among student outside of school time but are still related to students through students’ social gatherings and daily life outside of school time. Lastly, the general population risk factors include total COVID-19 infection case and general government policies for the public within the Victoria area.

Figure 2

Sample dataset for the COVID-19 infection simulation model. The dataset includes the daily number of infections in that age group, the method of acquisition, the local government area, the effective reproduction number (Rt), and the policy indicator, etc.

Figure 3

Incremental batch learning method for COVID-19 infection simulation model. Each batch included 14 consecutive days of records. The first batch of data was used to train model 1, while subsequent batches were used to update model parameters.

Figure 4

Gate recurrent unit (GRU) structure. The sigmoid function used in the reset gate and update gate of the GRU to decide the amount of information for forgetting and updating. The tanh function is used to calculate the new hidden state h_t−1.

Figure 5

Daily confirmed COVID-19 cases for age groups 0–9 and 10–19 in (a) 2020, (b) 2021, and (c) 2022. The blue line shows the daily cases for the 0–9 age group, while the orange line shows the daily cases for the 10–19 age group.

Figure 6

Mean Squared Error (MSE) for incremental batch learning method. The Mean Squared Error (MSE) represents the simulation error of the model for the age groups 0–9 and 10–19, while the training repetitions represent the number of repetitions of the training process used to determine the number of training iterations in the simulation model.

Figure 7

(a) Impact score for age group 0–9. (b) Impact score for age group 10–19. The impact score is calculated using the proposed comprehensive impact score review method based on the RNN-GRU model to decide the effectiveness of risk factors in preventing COVID-19 infection. The risk factors are listed in descending order within each risk dimension. The school-related risk factors (prevention strategies, immunisation status for the education sector) are listed first, followed by student-related community risk factors, and general population risk factors.

Figure 7

(a) Impact score for age group 0–9. (b) Impact score for age group 10–19. The impact score is calculated using the proposed comprehensive impact score review method based on the RNN-GRU model to decide the effectiveness of risk factors in preventing COVID-19 infection. The risk factors are listed in descending order within each risk dimension. The school-related risk factors (prevention strategies, immunisation status for the education sector) are listed first, followed by student-related community risk factors, and general population risk factors.