AI lends an ear to help understand schizophrenia

Credit: Unsplash+

A team from the UCL Institute for Neurology has crafted a novel approach using artificial intelligence (AI) to delve deeper into the study of schizophrenia, a mental health condition affecting approximately 24 million people globally.

Their research, spotlighted in PNAS, explores the potential of AI in detecting distinctive speech patterns in individuals diagnosed with this condition.

Bridging Technology and Psychiatry

Traditionally, psychiatric evaluations, including those for schizophrenia, have leaned heavily on discussions with patients and their acquaintances, marginally employing tests like blood analyses and brain scans.

This conventional method doesn’t lend itself to a comprehensive understanding of the underlying factors of mental illnesses nor does it significantly facilitate the tracking of treatment effects.

This new research incorporated AI language models to analyze the verbal expressions of individuals with schizophrenia.

Participants, comprising 26 individuals diagnosed with schizophrenia and a control group of 26, were tasked with verbal fluency activities, such as listing as many words as possible within a category or beginning with a specific letter within a five-minute timeframe.

AI’s Insight into Speech and Mental Health

The AI technology, educated on a substantial volume of internet text to comprehend word meanings similarly to humans, was used to determine if it could predict the words participants would use.

It emerged that words offered by the control group were more accurately predicted by the AI than those from the individuals with schizophrenia, especially those with more acute symptoms.

The science team theorizes that these differences might be linked to how the brain forms and stores “cognitive maps,” which help relate memories and ideas.

This hypothesis gained some traction when brain scanning to explore activity in areas responsible for developing and housing these cognitive maps also revealed differences.

Forging Ahead with AI in Mental Health Research

Dr. Matthew Nour, the lead researcher, acknowledged that only recently has the automated analysis of language become accessible to professionals in medicine and science due to advancements in AI language models like ChatGPT.

He highlights, “This work shows the potential of applying AI language models to psychiatry—a medical field intimately related to language and meaning.”

Schizophrenia, with symptoms that may manifest as hallucinations, delusions, and altered behavior, impacts over 685,000 individuals in the UK alone.

The researchers are eager to extend the application of this technology to larger and more varied patient samples and speech settings, aiming to assess its viability in a clinical context.

With a future-focused perspective, Dr. Nour expressed enthusiasm about the intersection of AI language models and mental health research.

“By combining state-of-the-art AI language models and brain scanning technology, we are beginning to uncover how meaning is constructed in the brain, and how this might go awry in psychiatric disorders.”

He envisions the deployment of these AI tools in clinical settings in the coming decade, should they prove to be safe and robust, marking an exciting era in neuroscience and mental health research where technology and psychiatry converge to deepen our understanding and improve the management of mental health conditions.

If you care about depression, please read studies that vegetarian diet may increase your depression risk, and Vitamin D could help reduce depression symptoms.

For more information about health, please see recent studies that ultra-processed foods may make you feel depressed, and these antioxidants could help reduce the risk of dementia.

The research findings can be found in PNAS.

Follow us on Twitter for more articles about this topic.

Copyright © 2023 Knowridge Science Report. All rights reserved.