The professor of Computer Science and Artificial Intelligence of the ULPGC, Luis Álvarez León, is carrying out a work in which he analyzes the real cases of those affected by coronavirus in Spain and in Canary Islands and compares them with two series of forecasts of the evolution of infections by coronavirus, to find the mathematical model that explains the evolution of the disease. This analysis is also carried out for the cases of China and South Korea, Italy and France.
With this mathematical model the influence of the implementation of drastic isolation measures in a pandemic is studied, taking into account the time required for those affected to be included in the official number of infected patients. This study includes daily conclusions from March 24 that change dynamically every day.
The professor emphasizes that in this work it is important spoint out that the forecasts shown may contain significant errors due to the small amount of data, that there were no precedents for models that took into account the effect of massive population confinement in the spread of a virus, and that the proposed model could not be sufficiently verified. Therefore, it is necessary to interpret the data with great caution and prudence.
For the Canary Islands and Spain, on Sunday 29 it is stated that "we have had good data with a reduction in the number of new cases in both Spain and the Canary Islands. According to the model, in the next few days the lateral displacement should be maintained in the number of new daily cases, with ups and downs before starting the downward trend. It will be necessary to be attentive to how Italy evolves. According to the model, In the next few days, Italy should start to drop significantly the number of new daily cases. "
If our pattern of behavior according to the model is similar to that of Italy, until April 4 or 6 it will continue to rise and fall in the number of new cases forming a "plateau". From then on, you should start there is a change in trend and the number of new cases should start to drop, first gently and then faster. In this sense it will be important to see how Italy evolves next week, since it would be his turn to start the descent phase. This forecast, therefore, could change with updates in the coming days.
So far, the work of the professor of the ULPGC shows that the model can adapt quite well to the evolution in the case of China. In the case of Spain, the number of samples after the isolation date may be too small to correctly calculate the model and the estimate changes quite a bit every day. In this case, it seems that more data is needed to stabilize the estimate.
LThe estimate for Italy is now quite stable, as is also the case with France, suggesting that the results obtained go in the right direction. In the case of China, the model seems to suggest that the isolation measures in the Asian country were slightly stronger than those in Italy and France, although the models obtained clearly indicate that the isolation measures in Italy and France are actually working and the The spread of the epidemic has slowed considerably (even though we are still seeing the number of people affected increase each day).
Keep in mind that it takes more than six days for most symptomatic people to be included in the infected dataset, so isolation measures take time to reduce the number of infected people.
In any case, the results obtained for China, Italy, France and Spain seem reasonable and, hopefully, promising.
The professor publishes in a site google the forecasts of this disease and, by way of conclusion, indicates that the Comparison of the results obtained for China, Italy, France and Spain seem reasonably similar and, hopefully, promising a positive evolution.