A synthetic intelligence app to read COVID-19 side tests helped decrease false effects in a new trial published today.
Published in Cell Reports Medicine, a team of researchers from the University of Birmingham, Durham University, and the University of Oxford tested whether a device learning a rule set can only accurately the effects of lateral antigen flow devices for COVID-19.
The LFD AI Consortium team worked in verification centres assisted by the UK Health Safety Agency and with fitness staff who performed self-checks to verify the AI application. More than 100,000 photographs were sent as part of the study, and the team found that the set of rules capable of increasing the sensitivity of the results, finding out between a true positive and a false negative, with an accuracy of 92% to 97. 6%.
The widespread use of lateral antigen flow devices has been a vital moment not only during the pandemic, but has also brought diagnostic tests for many more people in society. One of the disadvantages of LFD tests for Covid, pregnancy and any other long-term use is the “weak line” factor, where we cannot tell if it is positive or not.
The study looked at the option of employing device learning to take the guesswork out of checking weak lines, and we’re pleased to see that the app saw an increase in verification sensitivity, reducing the number of false negatives. The promise of this type of generation can be used in many applications, either to lessen uncertainty about the effects of control and to provide a very important supply to the visually impaired. “
Professor Camila Caiado, professor of statistics at Durham University and lead statistician of the project, said:
“The increased sensitivity and overall accuracy is significant and shows the prospect of this application to reduce the number of false negatives and infections in the long term. Importantly, the approach can also be seamlessly adapted to the evaluation of other virtual readers for lateral flow. type devices. “
University of Birmingham
The LFD AI Consortium. , (2022) Machine learning to discover the effects of side devices to test SARS-CoV-2 infection in asymptomatic populations. The mobile informs the médecine. doi. org/10. 1016/j. xcrm. 2022. 100784.
News-Medical. net – An AZoNetwork website
Owned and operated through AZoNetwork, © 2000-2022