An artificial intelligence model can determine with over 90% accuracy which gender a brain scan belongs to
Sanford Medicine researchers have developed an artificial intelligence model that enhances the understanding of sex-based variances in the human brain. Published in the Proceedings of the National Academy of Sciences on Feb. 19, this study enhances the understanding of the brain’s mechanisms and could help to treat neuropsychiatric disorders.
Brain development according to sex
Led by Vinod Menon, PhD, professor of psychiatry and behavioural sciences, the team behind the study aimed to address the crucial role that sex plays in brain development, ageing, and the manifestation of psychiatric conditions. “Identifying consistent sex differences in the healthy adult brain is a critical step toward understanding sex-specific vulnerabilities in neuropsychiatric disorders,” stated Menon.
Key factors from the study revealed that certain “hotspots” in the brain, such as the default mode network and the striatum and limbic network, played pivotal roles in distinguishing between male and female brains. These regions are involved in processes like self-referential information processing, learning, and reward response.
The research does not definitively determine whether these differences emerge early in life due to hormonal factors or societal influences. However, it provides evidence that sex-related disparities in brain organisation exist and may contribute to venerability to psychiatric disorders between genders.
Using artificial intelligence to determine the difference in brain activity
The researchers used artificial intelligence techniques to achieve these insights and analysed extensive brain imaging datasets. Using deep neural network models, they trained the AI to classify brain scans based on sex, uncovering subtle patterns that distinguish between male and female brains.
Researchers used expandable AI to identify the specific brain networks. They found that functional brain characteristics varied between sex’s and significantly impacted cognitive performance.
The study’s implications extend beyond understanding sex differences in the brain. The researchers developed sex-specific models of cognitive abilities, demonstrating that brain features associated with gender can predict cognitive performance. These findings underline the importance of considering sex-specific factors in neuropsychiatric research and clinical practice.
Sex-specific factors
The overall applicability of their AI models, can be utilised to investigate various aspects of brain connectivity and their relationship to cognitive abilities and behaviours.
The team plans to make their model publicly available, offering researchers a valuable tool for advancing our understanding of brain function and addressing challenges related to learning impairments and social functioning differences.
By recognising and studying sex differences, researchers aim to develop more effective strategies for diagnosing and treating neuropsychiatric disorders.
