England (Commonwealth Union) – Since the COVID-19 pandemic emerged in late-2019, the virus went from a disease we had limited knowledge on, to one of the most researched conditions in the last two years. Several new treatments and diagnostic tools to predict long term severe disease and vaccines that can be administered via inhalation are currently under trial.
Researchers have announced a machine-learning tool that may assist in marking patients who are most at risk of getting COVID-19 in hospital.
The device, a form of artificial intelligence (AI), predicted patients at high risk of getting COVID-19 in the research with 87 % accuracy. This technology was produced by scientists at Imperial College London and the Infection Prevention and Control unit at Imperial College Healthcare NHS Trust.
Scientists created the tool with routine hospital and patient data. The tool was trained to mark risk factors linked to COVID-19 infections like age, gender, contacts with other infectious patients, where beds were located and patient movement within the hospital.
Foreseeing the patients who are at risk of being infected in hospital may assist in preventing onward transmission to other patients and staff. Researchers believe the tool can be applied to other infections where certain patients can be at risk of developing in hospital, like Clostridium difficile.
Lead author of the study Ashleigh Myall from the Department of Mathematics at Imperial College London, said: “Throughout the COVID-19 pandemic some patients developed the infection during their hospital stay. There is a need to develop predictive models to identify patients who are most at risk of contracting COVID-19 and intervene to mitigate against poor patient outcomes”, adding that they can carry this on with usual measures to lower outbreaks and other transmissions.
The researchers hope to move forward with this framework for omicron strains of COVID-19, other infectious diseases and to learn how the framework could be integrated into existing guidelines.
