The researchers analyzed blood samples taken from people who are at high clinical risk for psychosis. These people were followed for several years to see who developed a psychotic disorder and who did not. After examining the proteins in the blood samples and using machine learning to analyze this data, the scientists were able to find patterns of proteins in the early blood samples that can predict who will develop a psychotic disorder and track their medical condition
Scientists have discovered that testing the levels of certain proteins in blood samples can predict whether a person at risk of psychosis is likely to develop a psychotic disorder years later.
The study is published in the current edition of JAMA-Psychiatry and was led by researchers from the RCSI University of Medicine and Health Sciences.
Based on certain criteria, such as mild or brief psychotic symptoms, some people are at clinical risk of developing a psychotic disorder, such as schizophrenia. However, only 20% to 30% of these people will actually experience the psychotic disorder.
The researchers analyzed blood samples taken from people who are at high clinical risk for psychosis. These people were followed for several years to see who developed a psychotic disorder and who did not.
After examining the proteins in the blood samples and using machine learning to analyze this data, the scientists were able to find protein patterns in the early blood samples that can predict who will develop a psychotic disorder and track their medical condition.
Many of these proteins are active when the body fights inflammation, which suggests that there are early changes in the immune system in people who will develop a psychotic disorder. The findings also show that it is possible to predict their results using blood samples taken several years in advance.
The most accurate test was based on the ten most predictive proteins. This combination correctly identified who will develop a psychotic disorder in the future in 93% of the cases and also correctly identified those who will not with an accuracy of 80%.
"Ideally, we want to prevent psychotic disorders, but this requires the ability to accurately identify who is most at risk," said Professor David Cotter, Professor of Molecular Psychiatry at RCSI and author of the paper.
"Our research shows that with the help of machine learning, analyzing protein levels in blood samples can predict who is truly at risk and who may benefit from preventive treatments. Now we need to study these markers in other people who are at high risk for psychosis to confirm these findings."
The university has filed a patent application and the research team is working to commercialize this research through a license or partnership with industry.
to the notice of the researchers
More of the topic in Hayadan:
