Home Header News Machine learning for identifying bats carrying Nipah virus

Machine learning for identifying bats carrying Nipah virus

Indian flying fox
The algorithm could identify bat species already known to carry Nipah virus with 83% accuracy
Nipah is a deadly virus that can be transmitted to humans from the body fluids of infected bats. When infected bats feed on fruits or date palms, they contaminate the fruits. Once someone gets infected, he or she can spread Nipah virus to others. The recent outbreak in Kerala, which follows one that occurred in 2018, has brought the focus back on bats which are known to host the virus.
Using machine learning – a form of artificial intelligence (AI) – a group of Indian and international scientists has identified bat species with the potential to host Nipah virus. The research team included scientists from Kerala Agricultural University.

Who can host Nipah virus?

Based on traits of bats to carry the virus, researchers have pointed out that more bat species in India may be reservoirs of Nipah than the only one confirmed so far. The researchers trained a machine learning algorithm with data of 48 traits of 523 bats found in Asia, Australia and Oceania. Their algorithm ‘learnt’ the associations between the characteristics of these bats and whether they tested positive for the virus.
The data included 31 bat species found in India and 11 of these species have been found to host Nipah virus in studies elsewhere. They had evidence of past or current infection in their blood, urine, or tissue. The algorithm could identify bat species already known to carry Nipah with 83% accuracy. It also identified six species in Asia, Australia, and Oceana that have traits which make them possible hosts of the virus. Four of these species also occur in India, two of are found in Kerala.
“Our work provides a list of species to guide early surveillance and should not be taken as a definitive list of reservoirs. A series of further studies are required to triangulate on the reservoir hosts that pose a risk to humans,” researchers have cautioned in their study published in the journal PLoS Neglected Tropical Diseases. The machine learning analysis is based on already published scientific studies globally, and no samples have been actually tested for this study. Still, the findings are important from public health point of view.