You are not logged in and only seeing 7 days of articles. Please sign up or login to view more
Thursday 14 January 2021

Sensyne launches AI app for lateral flow tests

Sensyne Health logoOxford-based clinical AI company, Sensyne Health, has announced the launch of a smartphone app to help automate the reading and analysis of lateral flow tests, such as those used to test for COVID-19.

Its MagnifEye software uses deep learning AI to automate the reading of test results, helping enhance accuracy by avoiding human error and reading results beyond the human visible spectrum. It also facilitates the use of advanced analytics to interrogate data across large populations, helping to enhance epidemiological insights.

Lateral flow tests were in the news again this week after the BMJ published an article calling for an urgent rethink in the UK Government's COVID-19 lateral flow test roll out. It highlighted recent analysis that revealed 60% of infected symptomless people in the pilot study conducted in Liverpool went undetected, including 33% of those with high viral loads. Earlier studies have demonstrated that the tests are ineffective at identifying those at very early or very late stages of infection. The tests are most effective when used to screen people at very regular frequency and in detecting cases with high viral load.  

The application of AI could help to improve the accuracy of reading test results and should enhance disease surveillance across large populations.

MagnifEye, which utilises Microsoft Azure, draws on the work Sensyne has been conducting on the application of AI to the analysis of medical images (see Sensyne makes progress despite the challenges and work back). It was developed in response to demand from commercial test manufacturers and healthcare providers and has potential across a wide range of applications in human and animal health, not just for COVID-19.

Posted by: Dale Peters

Tags: analytics   lifesciences   AI   machinelearning   healthcare   covid-19  

Twitter   Facebook   LinkedIn   Email article link

© TechMarketView LLP 2007-2021: Unauthorised reproduction prohibited see full Terms and Conditions.