Scientists at the Canadian University of Alberta have succeeded in devising a method using AI to accurately predict the likelihood of a person developing schizophrenia in people with a family history of mental disorder, according to a report in the journal Neuroscience.
University of Alberta researchers took another step forward in developing an artificial intelligence tool to predict schizophrenia by analyzing brain scans, and the tool was used to analyze fMRI images of 57 healthy first-degree relatives (siblings or children) of patients with schizophrenia, and accurately identified the 14 individuals. Who scored highest on the Self-Reporting Dissociative Personality Traits Scale.
Schizophrenia can cause delusions, hallucinations, disorganized speech, problems with thinking, and lack of motivation, and is usually treated with a combination of medication, psychotherapy and brain stimulation.Relatives of first-degree patients have up to 19% of the risk of developing schizophrenia during their lifetime, compared to the general risks that are reduced. About one%.
"Our evidence-based tool looks at the neural signature in the brain, with the potential to be more accurate than diagnosing through self-assessment of symptoms alone," said lead author Sunil Calmadi Vasu, chief machine learning specialist at the University of Alberta School of Medicine.
Kalmadi Vasu noted that the tool is designed to be a decision support tool and will not replace a diagnosis by a psychiatrist and also noted that while schizoid personality traits may make people more likely to develop psychosis, it is not certain that they will develop complete schizophrenia, indicating that the goal is The tool helps in early diagnosis, study of the schizophrenia process, and to help identify groups of symptoms.
"The next steps of the research will test the accuracy of the tool on non-family individuals with schizophrenia traits, and track individuals who were evaluated over time to see if they developed schizophrenia later in life," said Gulmadi Vasu.
Comments
Post a Comment