Mathematicians Develop a New Model for Predicting Epidemics Based on Precedents

COVID-19. Within less than a year, this word has managed to become a buzzword - alarming all eyes, and all ears. Most people are familiar with what it is, but figuring out how to tackle this unpredictable virus has been a lengthy process with many obstacles, given how rapidly it spread and how unexpected it was. This problem of overcoming COVID-19 may be combatted better in the future, thanks to the development of a new model known as the CBRR model, which outlines COVID-19 projections using machine learning. While this model can’t predict the future, it can help us understand what might happen in certain scenarios. Ultimately, this means that the model can help governments plan and act to achieve the best possible outcome [1].

This much-needed development will enable a better understanding of any future outbreaks. If the dynamics of a pandemic like COVID-19 were better understood earlier, we could have been more prepared: this could have prevented many deaths, health issues, economic issues, among many other things. COVID-19 has harmed the health and wellbeing of millions worldwide, with almost 2.5 million deaths attributed to the disease, particularly in countries such as India, Brazil, the UK, and the US [2]. Its spread has left national economies and businesses in an economically unstable position, and triggered huge shifts in stock markets. In addition, it has financially impacted the lives of many individuals, unemployment rates have increased, and many people have faced income cuts [2]. For example, unemployment rates in Canada have increased by 4.5% since COVID-19 hit [2]. Unfortunately, this virus disrupted many industries, but the spread of future epidemics and pandemics can be minimized by a new model developed by the scientists of the Intelligent Logistics Centre at St Petersburg University, in St Petersburg, Russia [1].

Since the start of the outbreak, these scientists have been developing the CBRR (Case-Based Rate Reasoning) model, which predicts the dynamics of epidemics, including that of COVID-19 [1]. The model is based heavily on the data of the dynamics of the epidemic in countries where the disease was recorded earlier [3].

According to one of the scientists involved, the CBRR model is a “paradigm of artificial intelligence and cognitive science that models the reasoning process as primarily memory based” [3]. In other words, it predicts the virus’s spread based on the solutions for similar problems in the past.

This model analyzes the dynamics of the virus in various places around the world. The work to form the model began in April-May 2020 when there were no statistics on the dynamics of  SARS-CoV-2, only for known viruses. As a result of the virus’s unknown nature, no previous models were applicable for forecasting its dynamics, and therefore it was necessary to develop a new approach and a new CBRR model. The COVID CBRR model uses precedents to predict how outbreaks of diseases such as COVID-19 will spread [4]. For example, it used data on the dynamics of the spread of the new coronavirus in countries that had outbreaks before Russia, the country where the model was originally intended to be used [1]. They managed to develop a program that forecasts weekly updates in COVID cases, by making predictions for 2-3 weeks to come.

In the model, values like the number of cases in a specified region are given the most importance, as they help to best understand the load level of the healthcare system, and predict the burden it will face in the future [1]. This ultimately results in a more calculated and accurate forecast of the virus’ future dynamics. Some factors taken into account for this forecast include periods of peak height in cases, increased cases, and the percentage growth in cases. It uses an iterative procedure, which derives products using previous values/precedents.

According to Professor Zakharov, one of the key points of the iterative procedure is the Epidemic Spreading Chain (ESC). This chain includes several countries ranked by the time they took to reach the same levels of the same specified and varying parameters [1]. He further noted that countries must take identical measures, like quarantine, self-isolation, and social distancing, for accurate results throughout nations. Countries like Italy and the UK were ultimately also included in the model, as they used similar measures.

The COVID-19 pandemic was sudden, with scientific developments being required promptly due to its unpredicted initial emergence. The CBRR model is still being worked on, and the data being used for it is limited - as it is being used specifically for Russia. Nonetheless, it can pave the way for similar technology in other countries. Despite the model’s many benefits, alternative models may arise in the future. As scientists and mathematicians continue to better understand COVID-19, more accurate precautionary measures can be implemented - which will only be beneficial in the long run.

References

[1]L. Jones, D. Palumbo and D. Brown, “Coronavirus: How the Pandemic Has Changed the World Economy,” BBC News, January 24, 2021. [Online]. Available: www.bbc.com/news/business-51706225. [Accessed 5 February 2021].

[2] J. Press, “Women, Youth Bear the Brunt of January Job Losses, Unemployment Rate Hits 9.4%.” CTVNews, February 5, 2021. Available: www.ctvnews.ca/business/women-youth-bear-the-brunt-of-january-job-losses-unemployment-rate-hits-9-4-1.5296869. [Accessed 5 February 2021].

[3] A. Aamodt and E. Plaza, “Case-based reasoning: Foundational issue, methodological variations, and system approaches,” AI Communications, vol. 7, no. 1, pp. 39-59, 1994. Available: http://dx.doi.org/10.3233/AIC-1994-7104. 

[4] St Petersburg State University, “Mathematicians Develop a New Model for Predicting Epidemics Based on Precedents,” Phys.org, December 9, 2020. [Online]. Available: https://phys.org/news/2020-12-mathematicians-epidemics-based.html. [Accessed 5 February 2021]. 

[5] St. Petersburg State University, “Mathematicians Develop a New Model for Predicting Epidemics Based on Precedents,” News Break, December 9, 2020. [Online] Available:  www.newsbreak.com/news/2121673532118/mathematicians-develop-a-new-model-for-predicting-epidemics-based-on-precedents. [Accessed 5 February 2021]. 

[6] Public Health Agency of Canada, “Government of Canada.” Gouvernement Du Canada, February 2021. [Online]. Available: www.canada.ca/en/public-health/services/diseases/coronavirus-disease-covid-19/epidemiological-economic-research-data/mathematical-modelling.html#w. [Accessed 3 Feb 2021].