(Combating New Coronary Pneumonia) Zhong Nanshan's team recommends that preventive measures such as "early screening" be implemented until the end of April
China News Network, Guangzhou, March 2 (Cai Minjie) The Guangzhou Respiratory Health Research Institute reprinted the "AME Scientific Time" release on its official public account on March 2 stating that Zhong Nanshan, an academician of the Chinese Academy of Engineering, recently published " A research article on the prediction of COVID-19 (new coronary pneumonia) outbreaks in China under public health intervention based on SEIR optimization model and AI.
The study integrated population migration data around January 23 and the latest COVID-19 epidemiological data into a classic epidemiological prediction model (SEIR) and derived the epidemic curve.
An artificial intelligence (AI) method trained based on SARS data in 2003 to predict the epidemic trend of this epidemic. It is speculated that the epidemic in China reached its peak in late February and flattened towards the end of April. If the implementation of the control measures is postponed for 5 days, the scale of the outbreak in Mainland China is expected to triple. If the control of Wuhan is reduced, Hubei may have a second outbreak in mid-March and continue into late April.
Among them, the optimized SEIR model and AI model predictions show that as of the end of April, there were 90,000 to 120,000 outbreaks in mainland China. This article reminds the public not to misunderstand. There is a certain gap between model predictions and reality. If the government continues to strictly implement control The policy will improve the diagnostic level and launch the use of drugs, and the scale of the epidemic will be greatly controlled.
In addition, the occurrence of the second peak in Hubei mentioned in this article is an assumption made under the circumstance of reduced control. According to the current situation, the Hubei Provincial Government will continue to maintain strict control and the possibility of a second peak is relatively small.
At present, the forecast results of this article are based on the control policies before February 9. The scale of the epidemic is less than that predicted by The Lancet and overseas scholars. The data in this article show that, as of the end of April, the peak of existing confirmed cases (non-cumulative diagnoses) in the mainland of China was not higher than 70,000, Hubei was not higher than 52,000, and Guangdong and Zhejiang were not higher than 1,200. The article's prediction range is closer to the real situation.
The article states that the SEIR optimization model can accurately predict the epidemic trend of new coronavirus. At the same time, an artificial intelligence prediction model has been successfully established and shows a similar trend to the SEIR model, confirming that the public health intervention measures implemented since January 23 have effectively controlled Epidemic development. Strict prevention and control measures such as "early screening" are best implemented until the end of April 2020. . (Finish)