Chinese researchers have discovered by employing big-data analysis that the COVID-19 pandemic in the United States might have started to spread around in September 2019, earlier than the officially announced date of its first confirmed case.

According to a new article published Wednesday as a preprint in ChinaXiv, a series of previous studies showed that the United States, Spain, France, Italy, Brazil and other countries had shown signs of being hit by the virus before its outbreak in China.

ChinaXiv is an online publishing service operated by the National Science Library, Chinese Academy of Sciences.

The article, titled "Dating the First Case of COVID-19 Epidemic from a Probabilistic Perspective," suggested that the qualitative and quantitative analysis of infectious diseases, done by combining mathematical models and artificial-intelligence technology, can reveal the epidemic law of infectious diseases.

The researchers set up an optimized model using the epidemic transmission model and big-data analysis method, and inferred the dates of the first infection cases in 12 northeastern U.S. states and in China's Wuhan City and Zhejiang Province, based on published data.

The result indicated that, for the 12 U.S. states, the possible dates of the first infection, with a probability of 50 percent, fall mostly between August and October 2019, while the earliest is April 26, 2019 on Rhode Island, and the latest is Nov. 30, 2019 in Delaware.

All of the dates indicated by the data are earlier than Jan. 20, 2020, the officially announced date of the first confirmed case in the United States, showing that the COVID-19 pandemic in the United States started to spread around September 2019 with a high confidence probability.

The result also showed that the date of the first COVID-19 case in Wuhan, with a probability of 50 percent, is Dec. 20, 2019, and the date for Zhejiang is Dec. 23, 2019. It infers that the COVID-19 in China is most likely to have started in late December 2019.

The article said this is consistent with the results of the epidemiological investigation, which proves that the calculation method is accurate and reliable.