The risk beyond elephants
Tuberculosis is a bacterial infection that attacks the lungs. According to the US Centers for Disease Control and Prevention, tuberculosis is one of the deadliest diseases in the world. Although uncommon in humans in the United States, 10.4 million people worldwide became ill from the disease in 2016, and there are 1.7 million TB-related deaths worldwide each year.
In humans and animals, most types of TB can be treated with antibiotics when the disease is caught early enough, but epidemiologists fear that a drug-resistant outbreak of the disease can cause a global pandemic.
Transmission of TB from animals to humans is rare but not unheard of. In 2013, seven people who had close contact with TB-infected elephants in Oregon tested positive for the disease.3
A better way to test for TB
A passionate data scientist and zoological expert has worked to develop a more accurate and less invasive method for identifying tuberculosis in elephants using neural networks, a type machine learning.
Currently, zoos and sanctuaries carry out annual TB tests with trunk washing on each elephant that is in their care. And new elephants in their herds are often tested or temporarily isolated to reduce the transmission of the disease.
New research from Sarah Harden, systems engineer at SAS, and Dr. Ramiro Isaza, professor of zoology at the University of Florida, analyzes 20 years of data on elephants in the United States. The research compares traditional methods of analysis, such as logistic regression and decision trees, with more advanced methods, such as neural networks and ensemble modeling.
According to Harden, the network models outperformed all other methods for identifying the probability of tuberculosis in individual elephants because this advanced method analyzes relational factors. These factors include elephant locations, herd dynamics, and social groups found within the elephant population.
Analysis of network variables shows the relationships and connections between the elephants and identifies which animals are most likely to be at risk of the disease.
“In many cases, the network model can detect TB before the standard test detects it,” Harden explains. Zookeepers could use this information to know which elephants to test, which ones to treat and which ones to isolate. Plus the analysis is less invasive, cheaper and takes less time.