diagnostics.ai (formerly Azure PCR) since 2014 has focused on solving patient safety issues caused by known problems with diagnostic testing. Many diagnostic issues are data-driven and resolving them requires new data-science based approaches. diagnostics.ai is focused on using innovative AI and machine-learning to improve diagnostic accuracy and cut costs – improving patient safety worldwide.
Many diagnostic tests require manual inputs and analysis, making them time consuming, expensive and leaving room for human error. diagnostics.ai’s initial focus is detection of infection, since misdiagnoses can lead to disease spread and evolution of antimicrobial resistant strains. Our first product, pcr.ai, enables automation of one of the world’s most popular methods of infection testing, – qPCR, where unexpected error rates as high as 20% (or more) have been documented. More information.
qPCR error rates
Studies have shown prevalent qPCR methods to have as high as a 20% error rate. In fact, a recent article in the Wall Street Journal called lab testing the “Wild West of Medicine.” In academic literature, global diagnostic rates for all test types are recorded as averaging 1 in 20 (National Academies of Sciences, Engineering, and Medicine (2015); Improving diagnosis in health care; The National Academies Press).
An NHS hospital, Edinburgh Royal Infirmary, recorded error rates of up to 20% (“mainly due to errors in manual analysis”) when using a semi-automated qPCR test for monitoring CMV infections in immunocompromised patients.
In comparison, results demonstrating that pcr.ai technology had no analysis errors when used with a semi-automated CMV test were published by NHS G&C hospital (Journal of Clinical Virology: Volume 70, Supplement 1 (Sept 2015)- abstracts 1585 and 1736 respectively).