AI identified that men and women have different brains

Stanford Medicine researchers have created a model that can successfully identify males and females just by looking at their brain scans.

Scientists have trained a deep neural network model to classify MRI brain imaging data by showing brain scans to an AI platform and tagging the male and female brains.

Interestingly, the model started noticing patterns, and when researchers tested the model on around 1,500 brain scans, it could tell with 90% accuracy which gender the brains belonged to.

Researchers used Explainable AI (XAI), a system that’s designed to explain the decision-making of artificial intelligence. It gave them insights into the “hotspots” on the images that most helped the model distinguish male brains from female ones. The areas that helped the AI to differentiate gender included the default mode network, a brain system that helps us to process self-referential information, such as one’s memories, actions, and plans, and the striatum and limbic network, which are involved in learning and how humans respond to rewards.

The team tried making another model that predicts how well people would perform on cognitive tasks, using brain features that differ between men and women. They created separate models for each gender: one predicted well for men but not women, and the other for women but not men. This suggests that brain differences between genders have a significant impact on behavior.

“These models worked really well because we successfully separated brain patterns between sexes,” said Vinod Menon, PhD, professor of psychiatry and behavioral sciences and director of the Stanford Cognitive and Systems Neuroscience Laboratory.

“That tells me that overlooking sex differences in brain organization could lead us to miss key factors underlying neuropsychiatric disorders,” he said.

While science knew the role of sex chromosomes in influencing hormone exposure during key life stages, such as early development and aging, they've struggled to pinpoint clear differences between male and female brains.

Brain structures look similar across genders, and previous research on how different brain regions function together hasn't consistently revealed distinct sex-related patterns.

The findings of the recent study might settle a longstanding debate on the existence of sex differences in the human brain. Understanding these differences could be crucial in tackling neuropsychiatric disorders that impact women and men differently.

The team applied their deep neural network model to study sex differences, but Menon says it could also help understand how different aspects of brain connections relate to various cognitive abilities or behaviors. They’re planning to make their findings and AI model publicly available for any researcher to use.

“Our AI models have very broad applicability,” Menon said. “A researcher could use our models to look for brain differences linked to learning impairments or social functioning differences, for instance – aspects we are keen to understand better to aid individuals in adapting to and surmounting these challenges.”

Vinod Menon told Cybernews, that the differences in brain activity patterns identified by AI are too subtle for humans to detect based on visual inspection. “The sophisticated algorithms used in our study can detect intricate patterns and associations within complex time series datasets, such as those obtained from functional brain imaging,” he said.