Since the beginning of the COVID-19 pandemic, various research teams have been working on prediction models using patient history and public health data to assess the severity of the condition of a person infected with the coronavirus and to try to eliminate the risk.
Over the months, experts have managed to identify several risk factors that increase the chances of someone dying from COVID-19. Now, a new study from the University of Copenhagen (Denmark) and collected by the journal Nature , has shown that artificial intelligence (AI) can help predict with 90.2% accuracy if someone will die of COVID-19 sooner or later. after becoming infected by evaluating some of these risk factors. Something that will serve to predict the number of patients in hospitals who will need a ventilator and determine who should be the first to receive the precious vaccine.
The results of the study also show that artificial intelligence can, once a patient is admitted to hospital with COVID-19, predict with 80% accuracy whether the person will need a respirator, something that could help relieve hospital pressure.
“We started working on the models to assist hospitals, as during the first wave they were afraid they didn’t have enough ventilators for intensive care patients. Our new findings could also be used to carefully identify who needs a vaccine, ”explains Mads Nielsen from the Department of Computer Science at the University of Copenhagen and leader of the work.
Who is more likely to die from COVID-19?
The machine learning model developed in the study is based on health data from 3,944 Danish COVID-19 patients collected from the UK Biobank. The model took into account several risk factors such as body mass index (BMI), gender and high blood pressure – the most weighted factors – and then the AI used the data to identify patterns and correlations with previous illnesses and contagion. of COVID-19 patients.
The data concludes that older men with high blood pressure are at higher risk than the rest.
“Our results show, unsurprisingly, that age and BMI are the most decisive parameters of the severity of severity of COVID-19 involvement in a person. But the probability of dying or ending up on a ventilator also increases. if you are a man, you have high blood pressure or neurological disease, “ explains Nielsen. “For those affected by one or more of these parameters, we have found that it may make sense to move them up the vaccine list, to avoid any risk of them becoming infected and eventually ending up on a respirator.”
Obviously, it is necessary to mention the limitations of the study; Among other things, the sample is relatively small, but it would still be useful in assisting and identifying patients who are most at risk and may serve as a potential tool in clinical settings in the future.
“We are working to achieve the goal of being able to predict the need for respirators five days in advance by offering the program access to health data for all COVID positives in the region,” says Nielsen. “AI can never replace a doctor’s evaluation, but it can help doctors and hospitals see many COVID-19 infected patients at once and set priorities.”
Referencia: Jimenez-Solem, E., Petersen, T.S., Hansen, C. et al. Developing and validating COVID-19 adverse outcome risk prediction models from a bi-national European cohort of 5594 patients. Sci Rep 11, 3246 (2021). https://doi.org/10.1038/s41598-021-81844-x