A study published in the journal JAMA Psychiatry has demonstrated that the testing of blood samples for the levels of certain proteins may help predict if an individual is at the risk of psychosis and if they develop any psychotic disorders in the future.
David Cotter, study author, said in a statement, "Our research has shown that, with help from machine learning, analysis of protein levels in blood samples can predict who is truly at risk and could possibly benefit from preventive treatments."
Using Machine Learning to Analyze Data
Based on certain criteria, such as mild or brief psychotic symptoms, some people are considered to be clinically at high risk of developing a psychotic disorder, such as schizophrenia. However, only 20 percent to 30 percent of these people will actually go on to develop a psychotic disorder. For the study, researchers analyzed blood samples taken from people clinically at a higher risk of psychosis.
These individuals were followed up for several years to see who did and did not develop a psychotic disorder. After assessing the proteins in blood samples and using machine learning to analyze this data, scientists were able to find patterns of proteins in the early blood samples that could predict who did and did not develop a psychotic disorder.
Potential to Prevent Psychotic Disorders
Many of these proteins are involved in inflammation, suggesting that there are early changes in the immune system in people who go on to develop a psychotic disorder. The findings also suggest that it is possible to predict their outcomes using blood samples taken several years in advance.
The most accurate test was based on the 10 most predictive proteins. It correctly identified those who would go on to develop a psychotic disorder in 93 percent of high-risk cases, and it correctly identified those who would not in 80 percent of the cases.
"Ideally, we would like to prevent psychotic disorders, but that requires being able to accurately identify who is most at risk," Cotter said."We now need to study these markers in other people at high risk of psychosis to confirm these findings," Cotter noted.
(With inputs from agencies)