Attempt to derive useful information to prevent recurrence of new coronavirus infection using AI

In Japan, the position of the new coronavirus infectious disease (COVID-19) in the Infectious Diseases Law has been moved to category 5 in May 2023 . However, the threat of the new coronavirus has not disappeared, and as of August 2023, tens of thousands of infected people are still occurring per week. The establishment of an effective treatment method is required for the convergence of COVID-19, but a research paper published in April 2023 states that AI can be used to prevent the recurrence of COVID-19. It has been reported that the necessary factors for finding 'drug combinations' have been discovered.

Frontiers | Learning from real world data about combinatorial treatment selection for COVID-19
https://doi.org/10.3389/frai.2023.1123285

Artificial Intelligence Uncovers the Best Drug Combos To Prevent COVID Recurrence
https://scitechdaily.com/artificial-intelligence-uncovers-the-best-drug-combos-to-prevent-covid-recurrence/

A research team by researchers at the pharmaceutical company Merck & Co. and the University of California selected AI-based COVID-19 treatments. According to the research team, about 30% of patients hospitalized with symptoms of COVID-19 in China tested positive for the new coronavirus again within 28 days after discharge. Based on this fact, the research team decided to focus on the recurrence rate rather than the mortality rate of COVID-19.

Fewer drugs are commonly used to treat COVID-19, but up to eight drugs have been prescribed in China since the early stages of the pandemic. For this reason, the research team was able to seek causal relationships between multiple drug combinations and patient characteristics by analyzing treatment data from China.

As a result of the research team analyzing treatment data from more than 400 people in China using machine learning, the optimal drug combination changes depending on the patient's 'age', 'systolic blood pressure', and ' blood CRP concentration '. It became clear.

According to the research team, the results of the research not only discovered ``the conditions for selecting the best drug for treating COVID-19,'' but also developed a method for extracting causal relationships from data with confounding factors. It is said that there is also a point.

In addition, research on using AI in the medical field is being conducted around the world, and in July 2023, research results were reported that the drug discovery process could be overwhelmingly shortened using AI.

It is reported that AI has derived a candidate molecule for an anti-aging drug that takes several weeks or more to investigate in a few minutes, the possibility of accelerating the drug discovery process - GIGAZINE