Hidden Disparities in Male and Female Brain Buildings Are Discovered by AI

Summary: Light matter is the subject of research using artificial intelligence to reveal unique cellular differences between men and women’s brains. These findings demonstrate how quickly AI may recognize sex-based mental patterns that are invisible to human eyes.

These differences, according to the research, could be used to improve mental ailments ‘ diagnostic devices and treatments. To provide detailed insights into neurological diseases, this study emphasizes the need for variety in mind studies.

Important Information:

    AI Accuracy: Artificial designs identified genetic intercourse in MRI imaging with 92 %- 98 % accuracy.

  1. White Matter Focus: Differences were found in the body’s light matter, essential for inter- local communication.
  2. Improved tests and treatments for conditions like autism and multiple sclerosis can be improved by understanding sex-based head differences.

Origin: NYU Langone

A new study reveals that the brains of men and women are organized at the cellular level in artificial intelligence ( AI ) computer programs that process MRI results. Light matter, which is mainly located in the innermost layer of the human brain and facilitates communication between regions, was the source of these variations.

Men and women are known to experience several sclerosis, autism spectrum disorder, nausea, and other mental problems at different prices and with varying symptoms.

So, it is important to develop better diagnostic tools and solutions in order to fully understand how biological sex affects the brain.

But, while head dimension, shape, and weight have been explored, scientists have just a partial picture of the brain’s structure at the biological level.

The new research, led by NYU Langone Health experts, analyzed hundreds of MRI brain imaging from 471 people and 560 people using an AI technique called machine learning.

Results demonstrated that computer programs could recognize patterns in construction and difficulty that were unobserved by humans and were able to distinguish between natural male and female brains.

Three different AI models were used to test the results ‘ ability to determine biological sex by examining their relative strengths when examining ties across multiple brain regions or by focusing on specific areas of white matter.

Our results provide a more in-depth understanding of the structure of a living, human mind, which may in turn provide new information into how many psychiatric and neurological disorders develop and how they can manifest in different ways in both men and women, according to study senior author and neuroradiologist Yvonne Lui, MD.

According to Lui, a teacher and evil chair for research in the NYU Grossman School of Medicine’s Department of Radiology, animal models and people tissue samples have been mostly used for studies of brain morphology.

Additionally, using statistical analyses of “hand-drawn” parts of curiosity, researchers had to make a lot of personal choices about the design, size, and location of the regions they chose, which raised questions the validity of some of these earlier findings. Such choices is probably skew the results, says Lui.

The authors claim that the new study’s findings, which were published online on May 14 in the journal  Scientific Reports, prevented that issue by using machine learning to analyze complete groups of images without having to go through a computer to check any particular location, which helped to remove people biases.

The squad began by providing AI programs with examples of brain scans from good men and women and by providing them with natural sex information.

These models eventually “learned” to distinguish biological sex on their own because they were designed to use complex statistical and mathematical techniques to become” smarter” as they gathered more data. Interestingly, the courses were restricted from using total head size and shape to produce their assessments, says Lui.

All of the models, according to the results, were able to accurately identify the sex of subjects who scan between 92 % and 98 % of the time. The systems were able to determine how quickly and precisely water may flow through brain tissue thanks to a number of factors.

Junbo Chen, MS, a doctoral candidate at NYU Tandon School of Engineering, said,” These results demonstrate the importance of diversity in studying diseases that arise in the human mind.

” If, as has been generally the case, men are used as a regular model for several disorders, researchers perhaps miss out on vital insight”, added study co- lead author Vara Lakshmi Bayanagari, MS, a graduate research assistant at NYU Tandon School of Engineering.

Bayanagari warns that the AI tools cannot reveal which sex was more likely to have which features, despite the ability to detect differences in brain-cell organization. She adds that the study only examined MRIs from cis-gendered men and women and classified sex based on genetic information.

The team next intends to research the evolution of sex-related brain structure changes over time in order to better understand environmental, hormonal, and social factors that might influence these changes.

Funding: Funding for the study was provided by the National Institutes of Health grants R01NS119767, R01NS131458, and P41EB017183, as well as by the United States Department of Defense grant W81XWH2010699.

In addition to Lui, Chen, and Bayanagari, other NYU Langone Health and NYU researchers involved in the study were Sohae Chung, PhD, and Yao Wang, PhD.

About this news about neuroscience and AI research

Author: Shira Polan
Source: NYU Langone
Contact: Shira Polan – NYU Langone
Image: The image is credited to Neuroscience News

Original Research: Open access.
Yvonne Lui and colleagues ‘” Deep Learning with Diffusion MRI as in vivo Microscope Reveals Sex-Related Differences in Human White Matter Microstructure” Scientific Reports


Deep Learning using an in-vivo microscope and diffusion MRI revealed sex-related differences in human white matter microstructure

In neuroscience studies, biological sex is a crucial factor because it has been demonstrated that there are differences between cognitive functions and neuropsychiatric disorders.

Although cellular-level microstructural differences that are related to sex have previously been found to be gross statistically related to macroscopic brain structures, such as cortical thickness and region size, are not well understood. This could provide insight into brain health and disease.

Understanding brain disorders and diseases that manifest differently in different sexes can be gained by studying these microstructural differences between men and women.

A crucial in-vivo, non-invasive method that opens up a window into brain tissue microstructure is diffusion MRI.

Our study uses volumetric diffusion MRI data to identify white matter regions that are most different between men and women and create multiple end-to-end classification models that accurately estimate a subject’s sex. 471 male and 560 female healthy subjects ( age range, 22–37&nbsp, years ) from the Human Connectome Project are included.

Fractional anisotropy, mean diffusivity and mean kurtosis are used to capture brain tissue microstructure characteristics.

To reduce bias caused by macroscopic anatomical differences like brain contour and size, diffusion parametric maps are registered to a standard template.

This study employ three major model architectures: 2D convolutional neural networks, 3D convolutional neural networks and Vision Transformer ( with self- supervised pertaining ).

Our findings demonstrate that all three models exhibit high sex classification performance across all diffusion metrics, which indicates that there are significant differences between male and female white matter tissue microstructures.

We also examine the relationship between short-range and long-range interactions by using complementary model architectures to understand the patterns of microstructural differences observed.

Which white matter regions are most important for sex classification are analyzed together with the Wilcoxon signed-rank test.

The results show that tissue microstructures exhibit sex-related differences in both local and global features and in longer-distance interactions.

Our highly consistent findings across different models provide new evidence that male and female brain cell-level tissue organization, particularly in the central white matter, is different.