PanDict to Enhance Pandemic Preparedness：A Multi-Model Framework for Transmission Dynamics, Variant Evolution, and Hospitalization Resource Planning

Abstract

Global epidemics, like COVID-19, have substantial impacts on almost all countries in multiple aspects, such as economy, hospitalization, lifestyle, etc1,2 . COVID-19 can spread to populations worldwide due, in part, to their high contagiousness, but more importantly, because of our inability to quickly address some of the most fundamental problems of a newly emerged virus: 1) How quickly will the virus spread? 2) Whether and under what conditions will new variants emerge?  3) How to arrange our resources accordingly? Since previous epidemic models were incapable of addressing these three most important questions, we developed the PanDict system, which can help address all three of the most essential problems discussed above. To elaborate, our model consists of three crucial parts, each tackling one of the three above-mentioned problems: 1) predicting the spread of the new virus in each local community and calculating its R0  value using our newly devised EPSEIRV model; 2) creating and using the SI3R model to simulate variant competition; 3) forecasting hospitalization deficiencies in each state and producing visual representations of the projected demand using our IHOV model. In contrast to other vague and incorrect predictions/models, our EPSEIRV model accurately predicted the spread of the Omicron variant of Sars-CoV-2 in the United States and South Africa prior to their peaks. Moreover, in January 2022, we concluded that the R0  value of Omicron is around 18.8. The high infection speeds of these viruses allow them to widely circulate in the population before vaccines are fully developed. Thus, there will be inevitable surges in the number of patients, which can potentially overwhelm unprepared hospitals, hence making the IHOV model especially imperative. In a nutshell, when a novel disease  emerges, the PanDict model can quickly and accurately predict how fast the disease spreads,   whether the disease will successfully mutate, and how to arrange hospitalization resources to   most efficiently mitigate suffering.    These crucial functions can apprise our users of where the  potential epidemic is heading and how to diminish its impact.  Furthermore, the PanDict model will allow hospitalization systems to be much more prepared for upcoming surges of patients, which would significantly reduce excess deaths and hospitalization deficiencies. The system also supports related departments or corporation plans with the EPSEIRV model and SI3R model during the contemporary epidemic.