Artificial intelligence is part of a new Covid-19 diagnostic method developed by the Federal University of Minas Gerais.
As a result of an investment of R$ 149,000.00 from the Research Support Foundation of Minas Gerais State, the Physics and Microbiology departments at the Federal University of Minas Gerais (UFMG) developed a new diagnostic method for Covid-19, and other diseases caused by viruses.
The platform uses artificial intelligence and optical spectroscopy to process diagnostic samples, comparing cells infected by the virus or not, and can also diagnose diseases such as dengue and zika. According to Professor Juan Carlos González Pérez, the method would be “accurate, fast, inexpensive and could be widely available in clinical analysis laboratories”.
Clinical tests of the method, which platform was called PoLiVirUS, showed an accuracy of 87%. Such success rate is comparable to the RT-PCR test, which would be the most effective method for Covid-19 diagnosing. Diagnostic indicators are often improved by the conduction of weekly new sets of tests.
More information is available, in Portuguese, in https://www.em.com.br/app/noticia/gerais/2020/07/16/interna_gerais,1167502/ufmg-utiliza-metodo-para-desenvolver-diagnostico-da-covid-19.shtml