The COVID-19 pandemic is increasing in many European countries, putting maximum pressure on hospitals around the world.
Innovative algorithms now help relieve pressure whenever hospitals face a new wave of COVID. Researchers at the University of Copenhagen have developed algorithms that can predict, among other things, the course of illness in COVID patients as to how many of them are likely or not needed. Intensive care Or ventilation.
This is important, for example, for staffing the entire Danish hospital, one of the study’s authors explains.
“For example, if you find that capacity issues occur after 5 days because too many beds are used in the Rigshospitalet, make a better plan and send the patient to a hospital with more space and staff. It can be diverted, so our algorithms have the potential to save lives, “explains Stephan Lorenzen, a post-doc in the School of Computer Science at the University of Copenhagen.
This algorithm uses individual patient data from the Sundheds platform (National Health Platform). This includes information about the patient’s gender, age, medication, body mass index, smoking status, blood pressure, and more.
This allows the algorithm to predict the number of patients requiring intensive care, such as on mechanical ventilation and continuous monitoring by nurses and doctors, within a time frame of 1 to 15 days.
Lorenzen, along with colleagues at the University of Copenhagen and researchers at Rigshospitalet and Bispebjerg Hospital New algorithm based on Health data From 42,526 Danish patients who were positive for coronavirus between March 2020 and May 2021.
Predict the number of patients in the intensive care unit with 90% accuracy
Traditionally, researchers have used regression models to predict COVID associations. hospital Admission fee. However, these models do not take into account individual medical history, age, gender, or other factors.
“Our algorithm is based on more detailed data than other models, which is more than 90% accurate in the number of patients admitted to the intensive care unit and those who require ventilation within 5 days. It means that you can predict with, “says Stephan Lorenzen.
In fact, this algorithm provides a very accurate prediction of the possible number of intensive care patients for up to 10 days.
“It provides better predictions than comparable models because it can more accurately map the potential need for ventilators and 24-hour intensive care for up to 10 days. Similar to the existing algorithmic model used for predictions. In addition, the accuracy is slightly higher than that. The course of the disease in COVID cases, “he details.
As a rule, the algorithm is ready to be introduced in Danish hospitals. As a result, researchers are about to begin discussions with relevant medical professionals.
“We have shown that the data can be used for a great many purposes. We are also fortunate to have a lot of health information available in Denmark. We hope that new algorithms will help hospitals avoid COVID overloads. I hope. The illness has struck, “Concludes Stephen Lorenzen.
What distinguishes a new algorithm from other algorithms
Most of the existing algorithms in this area do not consider an individual’s gender, age, or medical history. They look at the number of inpatient COVID patients in need of intensive care on a particular day. Based on this, along with mortality and new infection data, existing models try to predict the number of people hospitalized tomorrow.
“For example, a typical model cannot distinguish between a young model and a young model. senior citizen.. Whether there are five inpatients over the age of 80 or five 25-year-olds has a significant impact on the probability of hospitalization. Our new algorithm explains this, “says Stephan Lorenzen.
- new algorithm Use health data approved for use under Section 42d of the Danish Health Act.
- The data is processed by Computerome 2, a secure supercomputer for personal data, and is licensed by the Danish Patient Safety Authority, data owners, and other relevant authorities.
- The Danish Ethics Council approved the study and the regional executive committee approved the use of the data.
The study was published in Science report..
Stephan Sloth Lorenzen et al, Use of Machine Learning to Predict Intensive Care Unit Resource Use during a COVID-19 Pandemic in Denmark, Science report (2021). DOI: 10.1038 / s41598-021-98617-1
University of Copenhagen
Quote: New algorithm can save lives: https: //medicalxpress.com/news/2021-11-algorithm-covid-related-intensive-resource COVID related concentration obtained on November 24, 2021 Predict treatment room resource usage (November 24, 2021) .html
This document is subject to copyright. No part may be reproduced without written permission, except for fair transactions for personal investigation or research purposes. Content is provided for informational purposes only.
Predicts COVID-related intensive care unit resource use Source link Predicts COVID-related intensive care unit resource use
The post Predicts COVID-related intensive care unit resource use appeared first on California News Times.