Psychosis can be predicted before the onset using a machine-learning tool that can classify MRI brain scans into those who are healthy and those who are at risk of psychotic episodes. An international consortium, including researchers from the University of Tokyo, used the classifier to compare scans of more than 2,000 people from 21 global locations. About half of the participants were clinically identified as being at high risk of developing psychosis. Using the training data, the classifier was 85% accurate in distinguishing between those who were not at risk and those who subsequently experienced more psychotic symptoms. Using the new data, it was 73% correct. This tool may be helpful in future clinical settings, as most people who experience psychosis make a full recovery, earlier intervention usually leads to better outcomes with less negative impact on people’s lives.
Anyone can experience a psychotic episode, which usually involves confusion, hallucinations, or disorganized thinking. There is no single cause, but it can be triggered by illness or injury, trauma, drug or alcohol use, medications or genetic predisposition. Although it can be scary or unsettling, psychosis is treatable and most people recover. Because the most common age for a first episode is adolescence or early adulthood, when the brain and body are undergoing many changes, it can be difficult to identify youth who need help.
“Only 30% of clinically high-risk individuals later develop additional psychiatric symptoms, while the remaining 70% do not,” explains Shinsuke Koike, an associate professor at the University of Tokyo’s Graduate School of Arts and Sciences. “Therefore, clinicians need help identifying those who will develop psychotic symptoms using not only sub-clinical symptoms such as changes in thinking, behavior and emotions, but also some biological markers.”
The consortium of researchers worked together to develop a machine-learning tool that uses brain MRI scans to identify people at risk of psychosis before it starts. Previous studies using brain MRI have suggested that structural differences occur in the brain after the onset of psychosis. However, this is the first time differences have been identified in the brains of people who are at high risk but have not yet experienced psychosis.
The team from 21 different institutions in 15 different countries brought together a large and diverse group of adolescent and young adult participants. According to Koek, MRI studies of mental disorders can be challenging because the brain’s development and variations in MRI machines make it difficult to get very accurate, comparable results. Also, with young people, it can be difficult to distinguish between changes due to normal development and those due to mental illness.
“Different MRI models have different parameters that also affect the results,” Koeke explained. “Like cameras, different devices and shooting specifications produce different images of the same scene, in this case the participant’s brain. However, we were able to correct for these differences and create a classifier that is fine-tuned to predict the onset of psychosis.”
Participants were divided into three groups of people at clinical high risk: those who later developed psychosis; who did not develop psychosis; and a fourth group of uncertain follow-up status (total of 1,165 people for the three groups) and healthy controls for comparison (1,029 people). Using the scans, the researchers trained a machine-learning algorithm to detect patterns in the anatomy of the participants’ brains. From these four groups, the researchers used algorithms to classify participants into two main groups of interest: healthy controls and those at high risk who later developed excessive psychological symptoms.
In training, the tool was 85% accurate at classifying outcomes, while in a final test using new data it was 73% accurate at predicting which participants were at high risk of developing psychosis. Based on the results, the team considers that providing brain MRI scans to those clinically identified as high risk may be helpful in predicting future onset of psychosis.
“We still need to test whether the classifier will work well for new sets of data. Since some of the software we’ve used is best for a specific data set, we need to develop a classifier that can robustly classify MRIs from new sites and machines. , a challenge that a national brain science project in Japan, called Brain/Minds Beyond, is now taking on,” said Koike. “If we can do this successfully, we can develop more powerful classifiers for new data sets, which can be applied in real-life and routine clinical settings.”
Zhu, Y., etc (2024). Using brain structural neuroimaging measures to predict psychosis onset for clinical high-risk individuals. Molecular Psychiatry. doi.org/10.1038/s41380-024-02426-7.