Generally, the number of people who are positive for PCR test is the number of infected people, but there are many people who are not tested because they are asymptomatic infected people.Therefore, we analyzed the data obtained by correcting the number of deaths per day by the population of the country by machine learning that can be predicted for each region from the pattern of change.As a result, there was a marked regional difference of about 1 times between the Western and Asian regions in the spread of the new corona infection.There are various possible causes such as social, institutional, ethnic and genetic factors, but future research is needed.

　In addition, as a result of comparing the number of deaths due to infection with the status of PCR tests, the number of deaths in countries with a positive rate of less than 7% was only 15% in countries with a positive rate of higher.Countries with a positive rate of 2% or less showed a declining trend in daily deaths (Australia, Taiwan, China, South Korea).The positive rate in Tokyo on April 1 was 4%.

　Furthermore, it was found that the number of deaths increased several days after the increase in the number of positive PCR tests, but the period until the number of deaths increased was inversely proportional to the positive rate of the PCR test.Countries where positives were found and deaths increased immediately (short period of time) either overlooked early infections before symptoms due to inadequate PCR testing, or were too late to hospitalize severely ill. It is highly probable.

　From the above results, it is necessary to reduce the positive rate of PCR tests in order to reduce the number of deaths due to coronavirus infection, and for that purpose, the number of PCR tests will be widely expanded to asymptomatic people such as close contacts. We conclude that there is an urgent need.

Paper information:[Preprints] Global Comparison of Changes in the Number of Test-Positive Cases and Deathsby Coronavirus Infection (COVID-19) in the World