They create an AI that predicts the severity that COVID-19 patients can reach with 90% accuracy
The researchers trained the system with patient data to find patterns in their medical history. This AI could help prioritize vaccines and predict demand for respirators.
The fact that we have been suffering from this pandemic for so many months at a global level we only find one positive aspect, it has given researchers and health workers time to better understand the virus and know how to fight it more effectively. The following research project gives good proof of this.
Scientists from the University of Copenhagen have developed an artificial intelligence capable of predicting the clinical evolution of each person infected with coronavirus. Your results are 90% accurate, even when it comes to predict severe cases and deaths.
To train the system, it has been fed with health data of nearly 4,000 COVID-19 patients in Denmark. These parameters help artificial intelligence to find common patterns among all these people and to know what characteristics are frequent in patients who reach a greater severity.
These types of predictions seek to make more efficient use of the tools, materials and treatments that doctors have at their disposal. Would serve for give vaccines earlier to those with a potential risk of dying or to become seriously ill with the virus, managing to reduce the saturation of hospitals and the death rate.
The system has also achieved a 80% accuracy in predicting which patients will need a ventilator to overcome the disease. “We are working towards the goal of being able to predict the need for respirators five days in advance by providing computer access to health data for all COVID positives in the region“has explained the Professor Mads Nielsen from the University of Copenhagen.
We know that age and other factors are decisive in the degree of symptoms that a patient can reach. According to the study, men and people with high blood pressure or a neurological disease are at high risk.
“For those affected by one or more of these parameters, we have found that it may make sense to move them up the vaccine queue, to avoid any risk of them becoming infected and eventually ending up on a respirator.”Nielsen said. Lung diseases, dialysis, asthma, diabetes or heart disease are other important factors to take into account to take care of the weakest in the face of the pandemic.
The study was published this week in the Nature magazine and the artificial intelligence model is available online. Anyone can use this simulator to check the predictions of the system. However, the study of the University of Copenhagen acknowledges that the study has serious limitations, due in part to the number of patients that have been studied. With a larger sample the results would improve.
If you decide to try the simulator in this link, you should keep in mind that the result is not conclusive nor does it have the same validity as a doctor’s assessment. If you think you have been in contact with the virus and you feel unwell, call your nearest health center or the emergency number and follow the doctors’ recommendations.