Researchers from Emory University and Harvard University have developed a machine learning algorithm that can predict an individual’s likelihood for developing psychosis based on their speech patterns. Psychosis may be described as a “break with reality,” including hallucinations and delusions, which may arise from schizophrenia or other mental disorders. The researchers’ work, recently published in Nature Communications, demonstrates their tool can predict whether an at-risk person will develop psychosis with 90% accuracy. This is an exciting development for those who are at risk of psychosis, and can provide valuable medical information to aid in therapy and disease management.
Discover The World's MOST COMPREHENSIVE Mental Health Assessment Platform
Efficiently assess your patients for 80+ possible conditions with a single dynamic, intuitive mental health assessment. As low as $12 per patient per year.
Currently, clinicians use structured interviews and cognitive tests in order to assess an individual’s risk of psychosis. These methods demonstrate a roughly 80% accuracy, and only for patients at the highest risk levels of psychosis based on prior medical history. The researchers developed a new, quantifiable, and scalable machine learning approach to help improve accuracy and gather more valuable medical information for these patients.
The algorithm is based on an observation that there are specific, very subtle linguistic differences that occur in patients at risk of psychosis. These subtle changes include a reduction in sentence richness, and an increase in words related to sound. In a process dubbed “digital phenotyping,” the researchers are able to use the machine learning method to analyze a specific patient’s word choices and estimate the risk.
“In the clinical realm, we often lack precision,” says Neguine Rezaii, first author of the study. “We need more quantified, objective ways to measure subtle variables, such as those hidden within language usage.”
“This research is interesting not just for its potential to reveal more about mental illness, but for understanding how the mind works — how it puts ideas together,” says Phillip Wolff, senior author of the study. “Machine learning technology is advancing so rapidly that it’s giving us tools to data mine the human mind.”