Better Diagnosis of Schizophrenia with Artificial Intelligence

IBM scientists published a study in collaboration with the University of Alberta on the use of artificial intelligence and algorithms to better detect cases of schizophrenia, a rare disease still diagnosed from subjective factors, According to experts.

recent study published in the scientific journal Nature (in English) intends to push the limits of artificial intelligence to detect cases of schizophrenia.

The researchers were able to detect 74% of the patients participating in the study suffering from this disease. A similar proportion to doctors’ diagnoses at present.

According to the authors, doctors currently rely on subjective factors to perform their diagnosis, in contrast to this approach of computer-assisted psychiatry which offers tools to objectively evaluate patients.

“Right now, we do not know what’s wrong with the brain of a schizophrenic patient,” says Mina Gheiratmand of the University of Alberta and co-author of the study.

Scientists analyzed the brain activity of the participants’ brains using magnetic resonance imaging to “determine the differences between patients [and other study participants],” she continues.

The study also determined the degree of severity of schizophrenia-specific symptoms in affected patients.

Recommended For You


About the Author: Geoff Serka

Geoff is a former Senior Fellow at the Schuster Institute for Investigative Journalism at Brandeis University, is an award-winning travel, culture, and parenting writer. His writing has appeared in many of the nation’s most respected and credible publications, including the New York Times, the Washington Post, and on the cover of Smithsonian Magazine. A meticulous researcher who’s not afraid to be controversial, she is nationally known as a journalist who opens people’s eyes to the realities behind accepted practices in the care of children.