Artificial intelligence (AI) could play a crucial role in preventing future viral pandemics, claim scientists who have developed an early warning system utilizing machine learning. Researchers at the Scripps Research Institute, based in California, have trained the system to monitor the emergence and evolution of epidemic viruses, including variants of the SARS-CoV-2 virus responsible for COVID-19. According to the senior author of the paper, Prof. William Balch, this innovative system could offer an “unprecedented” approach in tracking viral pandemics. He stated, “There are rules of pandemic virus evolution that we have not understood, but can be discovered.”
The scientists revealed in a paper published in Cell Patterns that their AI system could have predicted new variants of COVID-19 weeks before they were officially recognized as threats by the World Health Organization (WHO). Prof. Balch emphasized the importance of considering not just the prominent variants but also the multitude of undesignated variants, referring to them as the “variant dark matter.” The AI successfully identified key variants from this “dark matter” that significantly impacted viral spread and mortality rates.
The AI system’s capabilities were demonstrated by its ability to track genetic changes in COVID-19 variants, as well as the virus’s response to various factors, including lockdowns, mask-wearing, new vaccines, increasing human immunity, and the competition between different variants. The scientists are optimistic that their findings indicate the potential for similar early warning systems to track the real-time evolution of future viral pandemics. Such systems could aid in predicting increases in infection rates, enabling the preparation of countermeasures like mask-wearing and healthcare service provisions.
Around 20% of healthcare organizations are currently using some form of AI.
AI technology in healthcare is primarily used by clinicians.
Furthermore, the AI system could prove valuable in the race to find treatments and vaccines during pandemics. It identified key COVID proteins and their roles in the pandemic’s evolution. Dr. Ben Calverley, co-first author of the study, highlighted the broad applications of the system, stating, “This system and its underlying technical methods have many possible future applications.”