Science & Technology

Which COVID-19 patients will eventually become infected with the ICU?This new algorithm can understand it

Researchers in California have created an algorithm that predicts the prognosis of COVID-19 in patients using only five data inputs. This is a significant improvement over the good old days of last year.

In the early days of the pandemic, healthcare professionals rationally collected patient data from different categories to take snapshots of patient illness and estimate prognosis. In today’s complex world of this infection, bedside tools can help physicians assess prognosis much faster.

COVID-19 is full of bad surprises and is sending a lot of people to hospitals and ICUs — well, you know the story. Hospital resources are maximally expanded in terms of both equipment and the person himself. The lack of available ICU beds has become a valuable nightmare for hospitals in a pandemic, so researchers sought tools to help predict ICU usage.

Few tools provide a real-time prognosis for a patient. It is very helpful for doctors to know how likely they are to enter the ICU after they have been admitted to the hospital. The new algorithm does just that. It is designed to give healthcare professionals some quick answers on the patient’s first admission.

Assad Oberai, a professor of Aerospace Mechanical Engineering, explained how biomarkers can tell how advanced COVID-19 is in patients.Oberai Said It is “a disease that affects many systems, not just one in the body. There are many problems at the same time. By focusing on these five functions, You can see how the disease is progressing. ”

Therefore, based on this, you can measure what is wrong with the patient and predict how the disease may progress. The indicators used by researchers to derive a patient’s prognosis are:

  • Dyspnea (patient’s respiratory rate and blood oxygen level);
  • Immune response (possible sepsis);
  • Cardiovascular system (possible blood clots);
  • And inflammation.

Because the predictor relies solely on quantitative data, it is less error-prone and subjective, and is scalable and applicable to all kinds of different scenarios. In addition, the model is simple and easy to use.

The team behind the algorithm is from the University of Southern California’s Viterbi School of Engineering and Keck School of Medicine.They announced their work in an article published in Science report so paper It is called “Machine Learning-based Predictor of COVID-19 Disease Severity”.

“Predictors of the need for intensive care and mechanical ventilation help healthcare systems plan the surge capacity of COVID-19,” the researchers explain in the study.

In a pandemic, the old but essential rule is to be positive rather than responsive. In this case, the author sought a proactive approach to resources such as ICU bends and ventilators.

“Given the urgency of resource allocation and optimization, we sought to identify patient-level clinical features at admission to predict the need for ICU care and ventilators in COVID-19 patients.”

Their study cohort consisted of 212 patients (123 males and 89 females) with an average age of 53 years, 74 of whom required intensive care at some point during their stay and 47 of whom were ventilators. Needed. “This aggressive approach is universally applicable to all emerging infectious diseases,” said Neha Nanda, medical director of infection prevention and antimicrobial management at Keck Medicine at USC.

However, to achieve this, the results need to be validated in a much larger cohort. So far, only the data obtained at the time of initial presentation was included as input to the predictive model. The need for ICU admission and mechanical ventilation was selected as a result at any time during hospitalization.

“The results presented in this study show that data obtained at or before or after admission to a care facility for COVID-19 patients can be used to accurately assess the need for critical care and mechanical ventilation. “The researchers conclude. ..

Which COVID-19 patients will eventually become infected with the ICU?This new algorithm can understand it

https://www.zmescience.com/science/which-covid-19-patients-end-up-in-the-icu-this-new-algorithm-could-figure-it-out/ Which COVID-19 patients will eventually become infected with the ICU?This new algorithm can understand it

Back to top button