Indian researchers use AI to map H5N1 bird flu spillover risk
As the H5N1 bird flu virus continues to evolve rapidly with the potential to pose a serious threat to human health, a team of Indian researchers has used an Artificial Intelligence (AI)-based model to decode how the virus could spill over from birds to humans and trigger sustained transmission.
The study, published in the journal BMC Public Health, uses BharatSim—an ultra-large-scale agent-based simulation framework originally developed for Covid-19 modelling—to map the sequential stages of a zoonotic spillover and the conditions under which a human outbreak could take off.
Two-step spillover risk identified
Researchers Philip Cherian and Gautam I. Menon from the Department of Physics at Ashoka University said the model captures a two-step process of outbreak initiation.
“We modelled the possibility of initial spillover events of H5N1 from birds to humans, followed by sustained human-to-human transmission,” the authors noted in the paper.
The model shows how key epidemiological parameters governing early transmission can be calibrated using data on the distribution of primary and secondary cases, offering insights into how outbreaks can be detected and controlled at an early stage.
Asia seen as likely hotspot
Avian influenza H5N1, which first emerged in China in the late 1990s, has occasionally infected humans over the years. With South and South-East Asia home to the world’s fastest-growing poultry markets, the region is considered the most likely location for an initial large-scale outbreak.
According to the World Health Organization (WHO), 990 human cases of H5N1 were reported across 25 countries between 2003 and August 2025, including 475 deaths—indicating a fatality rate of around 48 per cent.
Early action key to containment
The AI-driven model found that culling infected birds remains the most effective way to curb outbreaks, whether on farms or in wet markets—provided no primary human infection has already occurred.
The study also showed that if a primary human case emerges, isolating the patient and quarantining household contacts can still limit onward infections. However, once tertiary transmission begins, controlling the outbreak becomes extremely difficult without drastic interventions.
“Once tertiary contacts are infected, establishing control becomes impossible unless far more stringent measures are applied, including a total lockdown,” the researchers warned.
Lockdowns last resort
The researchers stressed that timely interventions during the earliest phase of an outbreak make the greatest difference.
“Once community transmission takes over, cruder public-health measures such as lockdowns, compulsory masking, and large-scale vaccination drives are the only options left,” they said.
The study highlights how advanced computational models like BharatSim can help policymakers test interventions in real time and improve understanding of the epidemiology of emerging infectious diseases before they spiral out of control.
For all the latest updates, download PGurus App.
