[ad_1]
While it is still very difficult for scientists to assess the evolution of coronavirus symptoms in patients, perhaps an algorithm could help them. Especially in the severe forms.
As vaccine announcements have been chained in recent weeks, science continues to work to improve the care of coronavirus-infected patients. And it could also be that start-up from La Rochelle put your finger on something. This is the medical bioinformatics company Numa Health. The latter would have created an algorithm capable of predicting severe forms of Covid-19.
95% reliable
Health Numea carried out a study in collaboration with the La Rochelle hospital. He took into account the cases of 50 patients who went through this between March and June. ” The software predictions were compared with the actual fate of the patients. “Explained Dr. David Chalvet, CEO of Numa Health, a France Blue.
And the least that can be said is that it turned out to be a great success.
” Result: for patients at low risk of degradation, thethe algorithm was good in 95% of cases ; for the high-risk patients, in 70% of cases. That is, patients entered the ICU or died within the next five days as predicted by the software. “
How does it work ?
The calculations used by the algorithm are based on a blood analysis. ” The idea is to use a simple blood test to determine which patients are at risk of getting worse because upon arrival at the ER, all markers are not necessarily red but the software can predict that they will become Added Dr. David Chalvet.
These analyzes are also associated with a file patient lifestyle questionnaire. All combined, this would allow caregivers to assess the severity of cases people arriving at the emergency room and performing quick sorting.
Towards commercialization
In a statement, Numea Health explains that the study will be applied to more cases to verify its effectiveness. The results will then be published in a medical journal and the algorithm could be inserted on the market at the end of 2021.
This algorithm could also be applied to other diseases.
Source link