AI-driven saliva test detects COVID-19 with high accuracy

A group of scientists at the UFU has developed a novel diagnostic system that accurately detects the SARS-CoV-2 virus in saliva. This is reported by Metropoles, a partner of TV BRICS. 

The approach combines electrochemical biosensors with two artificial-intelligence algorithms, marking a significant advance in portable infectious-disease testing. The technology builds on established principles used in blood-glucose monitoring: a biosensor analyses saliva for molecular markers of infection. 

What sets the UFU system apart is its deployment of dual AI models to refine detection. First, researchers identified a peptide naturally present in saliva that binds strongly to the SARS-CoV-2 virus. They then used one algorithm to optimise these peptides, enhancing the biosensor’s sensitivity. Then, the second algorithm is applied to interpret the resulting electrochemical signals, boosting diagnostic accuracy.

“This is the first time that AI-driven algorithms have been applied to peptide-based salivary detection of COVID-19,” said A. Sabino-Silva, coordinator at Federal University of Uberlandia. 

With the help of artificial intelligence, these molecules have been optimized to improve the accuracy of the biosensor in detecting the virus.