Center of Biomedical Physics and Information Technology
Future prediction (without the impact of the omicron variant): Update
The difference from the result of prediction up to November is that the prevention effect of vaccination was corrected into the reported value for Japanese people (advisory board meeting of Ministry of Health, Labour and Welfare (MHLW)). The implementation of countermeasures against infection is assumed.
Change in the number of new positive cases depending on the timing of the start of spread of the infection of the omicron variant
Calculated assuming that behavior control would start from February, referring to the flows of people of the previous year (black dotted line).
The implementation of countermeasures against infection is assumed. *The standard of flows of people is the median value of each day of the week for the five weeks
from January 3 to February 6, 2020
The timing of the third vaccination is fixed to eight months or longer after the second vaccination
A possibility is suggested that if the spread of infection of the omicron variant continued during the year-end and new-year holidays, the peak may increase due to the impact of behaviors during the vacation
5 (2) Factors of the decrease in infection (with the 5th wave as an example)
-There is a possibility that the decrease in the number of new positive cases appeared to be drastic due to (1) an end to the temporary increase in the effective reproduction number associated with consecutive holidays in the expansion period and (2) the infection-prevention effect of vaccination in addition to the decrease in contact opportunities.
- Used machine learning to virtually reproduce cases where (1) vaccination did not proceed from July 11 and onward, where (2) behaviors do not change, and where (3) there is no natural infection, and approximated the relationships based on the effective reproduction number between July 23 and September 20, 2021.
Learning up to April 2021. After that, actual data is used as much as possible for reproduction.
* Equivalent flows of people to the average value between January 6 and February 6, 2020.
** Twitter data is virtually set based on past data. Habits before the coronavirus catastrophe such as having a meal together without a mask are not considered.
***The number of asymptomatic infected people is assumed to be 4 times larger than the number of positive cases (approximate).
Effect to reduce the effective reproduction number
(Approximation: July 23 to September 20, 2021)
The table shows the average values of one week with a big reduction effect.
Reduction effect = [-(Rt(Prediction of the observed value)-Rt(Each condition))/Rt(Each condition)]
Rt: Effective reproduction number
*The impact of the long vacation is limited to August 1 to 4 (equivalent to July 22 to 25)
*Estimated based on the case of London (material of the AI periodic meeting of the Cabinet Secretariat/Hirata September 27)
***The weather conditions are the temperature and humidity only. The weather is not considered.
Modified the report material of the periodic meeting for the COVID-AI simulation of the Cabinet Secretariat (December 14, 2021)
5 (3) Related material: Relationship between the nighttime retention population and behavioral changes
(Learning up to April 15, 2021 and estimate for the following period for Tokyo)
Infectability of the virus
(Relative effective reproduction number)
The nighttime retention population is not necessarily an optimal indicator for the estimation of the number of new positive cases even if it is related to the number of new positive cases. The relationships with keywords such as karaoke, drinking party (not shown), etc. may be surrogates representing risk behaviors in general.
*Prepared using the Twitter data provided from NTT DATA by Toyoda Lab, Institute of Industrial Science, The University of Tokyo. (Excerpted from the material of the periodic meeting for the COVID-AI simulation of the Cabinet Secretariat (December 7, 2021))
