Artificial intelligence to predict the survival of covid patients in the ICU

Health systems around the world are struggling to care for the vast number of severe patients with covid who need special medical attention, especially if they are identified as high risk.

In the intensive care units scales or forecasting tools are often used, such as SOFA or Apache II, to predict the outcome of patients based on several parameters, but its reliability is limited in the case of covid-19.

Now, European scientists have shown that blood samples of a patient seriously ill due to this disease, specifically the proteins in their blood plasma, can be analyzed with a machine learning model predict weeks in advance whether the person will survive or not. The results are published in the open access journal Digital Health PLOS.

“Our study shows that a combination of proteomic markers, combined in a risk prediction model based on artificial intelligence, can very well predict the likelihood that an individual patient will die or survive Covid,” says co-author Florian Kurth Charité University Hospital in Berlin (Germany).

“Furthermore,” he adds, “the prediction of proteomic risk was much better than the prognosis derived from risk assessment scores commonly used in clinical care.”

Look for key blood proteins

To conduct the research, the authors began by studying the levels of 321 proteins in blood samples collected in 349 moments or time points in 50 critically ill COVID-19 patients who were being treated, with mechanical ventilation, in two health centers in Germany and Austria.

Then used the machine learning to find associations between measured proteins and survival of sick people. From the cohort or study group analyzed, 15 patients died and the mean time from admission to death was 28 days. For those who survived, the average hospital stay was 63 days.

The proteins with the greatest relevance in the prediction model belong to the coagulation and complement systems, and both are known to be especially important for the pathophysiology and severity of covid-19.

Read Also:  Europe is not prepared for climate threats

Florian Kurth (Charite)

Using blood tests, the researchers identified 14 proteins (such as alpha-2 macroglobulin, APOC3, GPLD1, various serpins…), whose measurements changed over time and moved in opposite directions on the graphs, depending on whether or not patients survived in intensive care.

“Interestingly, the plasma levels of all these proteins had already been altered by the disease, depending on the severity, which makes us confident in our findings”, says Kurth, who explains: “The proteins with greater relevance in the prediction model belong to the of coagulation and the so-called cascade or plugin system (a component of the immune response). Both are known to be especially important to the pathophysiology and severity of COVID-19.”

Survival Forecast

The team then developed their machine learning model to predict the survival of a single temporal measurement of the relevant proteins, and tested in Austria with 24 patients in critical condition because of the coronavirus.

For this group, the model demonstrated a high predictive power, correctly predicting 18 of the 19 patients who survived and five of the five people who died.

The researchers conclude that blood protein tests, once validated in larger cohorts, can be useful both to identify patients at higher risk of mortality and to better understand the disease and verify if a particular treatment changes the forecast when applied to individual cases.

“Now we want to see if we can transfer this methodology of research facilities for an everyday environment, to a standard clinical measurement laboratory, as well as evaluating the method in larger patient groups, and possibly for other diseases as well,” says Kurth.

Rights: Creative Commons.