Researchers at the University of Campinas (UNICAMP) in Brazil have developed a platform that combines mass spectrometry with an artificial intelligence algorithm to diagnose several diseases with a high degree of precision using metabolic markers. The researchers tested their platform to diagnose Zika, with accuracy exceeding 95%, according to findings published in Frontiers in Bioengineering and Biotechnology.
To develop and validate the platform, the researchers obtained blood samples from 203 patients, 82 of which were diagnosed with Zika by the real-time polymerase chain reaction (RT-PCR), currently the standard method of Zika diagnosis. Of those who had not been diagnosed with Zika, approximately half had symptoms that resembled the disease such as fever, joint pain, conjunctivitis and rash. The others displayed no symptoms and tested negative or were diagnosed with dengue.
All samples were analyzed in a mass spectrometer and approximately 10,000 different serum molecules were identified, some of which were produced by Zika and by the patient’s immune system in response to infection.
This data was then fed into a computer program running a random-forest machine learning algorithm able to perform classification, prediction, decision making, modeling, and more. The algorithm is able to return a set of metabolic markers to identify patients infected with Zika or the disease of interest. In the case of Zika, a panel of 42 biomarkers was established as being key to identifying the virus, 12 of which were considered “highly prevalent”.
"In this platform, it isn't important to know a lot individually about each of the molecules that serve as markers of the infection. It's the set that matters and that will tell us with a high level of accuracy whether we're looking at Zika. Moreover, even if the virus mutates, the program adapts and changes too. It's not a static methodology," said principal investigator, Rodrigo Ramos Catharino.
The team plans to make the platform available in the cloud so that it can be downloaded to other mass spectrometers.