Artificial intelligence helps identify types of prolonged COVID

  • According to WHO data, Worldwide, more than 660 million cases have been detected as of January 10, 2023.

  • As of January 10, 2023, 6,690,473 deaths have been reported around the world.

  • According to the WHO, in the last 24 hours more than 200,000 cases have been reported globally.

Research led by Justin Reese, from the Lawrence Berkeley National Laboratory (Berkeley, California), and Peter Robinson, from the Jackson Laboratory for Genomic Medicine (Farmington, Connecticut), has led to the development of an Artificial Intelligence tool to define the different types of prolonged COVID.

The artificial intelligence tool, described in a Investigation article published in eBioMedicine, analyzed the electronic medical records of people diagnosed with prolonged COVID.

According to the article, after analyzing the electronic medical records of 6,469 people with a confirmed diagnosis of COVID-19 and a subsequent diagnosis of long-term COVID, the team discovered six subtypes, or symptom groups, of long-term COVID. These include groups with distinct pulmonary, cardiovascular, and neuropsychiatric abnormalities, and a group associated with severe disease and increased mortality.

They also looked at relationships between long-term COVID symptoms and pre-existing illnesses, age, and other demographic factors. In addition, the Artificial Intelligence tool was able to demonstrate that the identified clusters were generalizable in different hospital systems.

“We compare all the symptoms of the pair of patients in this way and get a score of how similar the two long-COVID patients are. We can then perform unsupervised machine learning on these scores to find different subtypes of long-term COVID.” says Justin Reese in a Press release.

According to the study authors, this ability to determine groups among patients with chronic COVID provides a basis for classifying subgroups for treatments or therapies. In addition, the machine learning method self-adapts to different logging systems. This means that researchers can collect data from a wide range of medical centers.

Related notes:

Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.