Covid-19新冠肺炎的傳播干預及預測

Intervention and prediction of the spread of Covid-19 infectious disease

賴榮斌(Jung-Pin Lai)

指導教授 ： 白炳豐

國立暨南國際大學/管理學院/新興產業策略與發展博士學位學程/博士(2021年)

https://doi.org/10.6837/ncnu202200013

摘要

許多研究結果已經表明高疫苗接種率和遵守公共衛生預防措施對於控制Covid-19疫情流行並防止未來住院趨勢和死亡人數激增至關重要。實施公共政策的決策者可以通過加速疫苗接種增加覆蓋率來擴大群體免疫和實施各項非藥物干預措施 (NPIs)，例如隔離干預、關閉學校及設施、身體距離限制的及戴口罩等措施來控制疫情擴散，其中各國疫苗接種政策的實施亦是抑制疫情的重要關鍵。在本研究中使用隔室模型(Compartment model)中的感染數據用以模擬疫情傳播的狀態並獲取感染人數數據，預測模型使用長短期記憶（Long Short-Term Memory，LSTM）用以進行短期預測，訓練模型時使用基因演算法（Genetic Algorithm, GA）進行超參數優化。在感染人數預測中並進一步探討不同國家在政策干預下的情況，用以了解實施政策管制程度以及疫苗接種實施政策在不同國家中疾病傳播效果。本研究結果表明，所提出的GA-LSTM模型預測算法對於Covid-19疾病傳播在預測國家未來感染人數的可用性，誤差評估呈現精確的表現；另在透過非藥物干預措施(NPIs)及疫苗政策進行感染預測方面，研究結果亦顯示這是一個可靠且可執行的預測模型工具，政府管理者可透過此預測方法對疫情管制政策的實施進行決策評估，且可以即時依所在地需求進行政策調整。

關鍵字

Covid-19 ； 隔室模型 ； 預測 ； 長短期記憶模型 ； 基因演算法

並列摘要

Many research results have shown that high vaccination rates and compliance with public health precautions are essential to control the Covid-19 epidemic and prevent future hospitalization trends and a surge in deaths. Decision makers implementing public policies can expand herd immunity and implement various non-pharmaceutical interventions (NPIs) by accelerating vaccination and increasing coverage, such as isolation interventions, closing schools and facilities, restricting physical distance, and wearing masks to control The spread of the epidemic, and the implementation of vaccination policies in various countries is also an important key to curbing the epidemic. In this study, the infection data in the Compartment model is used to simulate the spread of the epidemic and to obtain data on the number of infected people. The prediction model uses Long Short-Term Memory (LSTM) to make short-term predictions. When training the model, genetic algorithm (GA) is used for hyper-parameter optimization. In the prediction of the number of infections, we will further explore the situation of different countries under policy intervention to understand the degree of policy control and the effect of disease transmission in different countries by the implementation of vaccination policies. The results of this study show that the proposed GA-LSTM model prediction algorithm has an accurate performance in predicting the availability of the Covid-19 disease spread in predicting the number of people infected in the country in the future, and the error assessment shows an accurate performance; in addition, it is carried out through non-pharmaceutical interventions (NPIs) and vaccine policies. In terms of infection prediction, the research results also show that this is a reliable and executable predictive model tool. Government managers can use this predictive method to make decision-making assessments on the implementation of epidemic control policies, and they can make immediate policy adjustments based on local needs.

並列關鍵字

Covid-19 ； Compartment model ； Forecasting ； LSTM ； Genetic Algorithm

參考文獻

1. Askitas, N.; Tatsiramos, K.; Verheyden, B. Estimating Worldwide Effects of Non-Pharmaceutical Interventions on COVID-19 Incidence and Population Mobility Patterns Using a Multiple-Event Study. Sci. Rep. 2021, 11.

2. Gumel, A.B.; Iboi, E.A.; Ngonghala, C.N.; Ngwa, G.A. Toward achieving a vaccine-derived herd immunity threshold for COVID-19 in the U.S. Front. Public Health 2021, 9, 709369.

3. Baden, L.R.; El Sahly, H.M.; Essink, B.; Kotloff, K.; Frey, S.; Novak, R.; Diemert, D.; Spector, S.A.; Rouphael, N.; Creech, C.B.; et al. Efficacy and Safety of the mRNA-1273 SARS-CoV-2 Vaccine. N. Engl. J. Med. 2020, 384, 403–416.

4. Bartsch, S.M.; O’Shea, K.J.; Ferguson, M.C.; Bottazzi, M.E.; Wedlock, P.T.; Strych, U.; McKinnell, J.A.; Siegmund, S.S.; Cox, S.N.; Hotez, P.J.; et al. Vaccine efficacy needed for a COVID-19 coronavirus vaccine to prevent or stop an epidemic as the sole intervention. Am. J. Prev. Med. 2020, 59, 493–503.

5. Jentsch, P.C.; Anand, M.; Bauch, C.T. Prioritising COVID-19 vaccination in changing social and epidemiological landscapes: A mathematical modelling study. Lancet Infect. Dis. 2021.