Brain organization differs between boys and girls with autism, according to a new study from the Stanford University School of Medicine.
The differences, identified by analyzing hundreds of brain scans with artificial intelligence techniques, were autism-specific and not found in typically developing boys and girls. The research helps explain why autism symptoms differ between genders and could pave the way for better diagnoses for girls, scientists say.
Autism is a developmental disorder with a spectrum of severity. Affected children have social and communication deficits, show restricted interests and display repetitive behaviors. The original description of autism, published in 1943 by Leo Kanner, MD, was biased in favor of male patients. The disorder is diagnosed in four times as many boys as girls, and most autism research has focused on men.
When a condition is described in a biased way, diagnostic methods are biased. This study suggests that we need to think differently.”
Kaustubh Supekar, PhD, lead author of the study, clinical assistant professor of psychiatry and behavioral sciences
The study was published online February 15 in The British Journal of Psychiatry.
“We detected significant differences between the brains of autistic boys and girls and obtained individualized predictions of clinical symptoms in girls,” said study lead author Vinod Menon, PhD, professor of psychiatry and behavioral sciences and Rachael L. and Walter F. Nichols, MD, professor. “We know that symptom camouflage is a major challenge in diagnosing autism in girls, leading to diagnostic and treatment delays.”
Girls with autism generally have fewer overt repetitive behaviors than boys, which may contribute to diagnostic delays, the researchers said.
“Knowing that men and women do not present the same way, both behaviorally and neurologically, is very compelling,” said Lawrence Fung, MD, PhD, assistant professor of psychiatry and behavioral sciences, who was not one of the authors of the study.
Fung treats people with autism at Stanford Children’s Health, including girls and women with delayed diagnoses. Many autism treatments work best during the preschool years, when the brain’s motor and language centers are developing, he noted.
“If treatments can be done at the right time, it makes a big, big difference: for example, children with autism who receive early language intervention will have a better chance of developing language like everyone else and won’t have to keep playing catch-up as they grow,” Fung said. “If a child can’t articulate well, they fall behind in a lot of different areas. The consequences are really serious if he doesn’t get an early diagnosis.”
New statistical methods reveal the differences
The study analyzed functional magnetic resonance imaging brain scans of 773 children with autism -; 637 boys and 136 girls. Amassing enough data to include a significant number of girls in the study has been a challenge, Supekar said, noting that the small number of girls historically included in autism research has been a barrier to learning more about them. . The research team relied on data collected at Stanford and public databases containing brain scans from research sites around the world.
The preponderance of boys in brain scan databases has also posed a mathematical challenge: standard statistical methods used to find differences between groups require groups to be approximately equal in size. These methods, which underpin machine learning techniques in which algorithms can be trained to find patterns in very large and complex data sets, cannot scale to a real-world situation where a group is four times larger than the other.
“When I tried to identify the differences [with traditional methods]the algorithm would tell me that every brain is an autistic male,” Supekar said. “It was overfitting and didn’t distinguish between autistic males and females.”
Supekar discussed the issue with Tengyu Ma, PhD, assistant professor of computer science and statistics at Stanford and co-author of the study. Ma had recently developed a method that could reliably compare complex data sets, such as brain scans, from different sized groups. The new technique provided the breakthrough scientists needed.
“We were lucky that this new statistical approach was developed at Stanford,” Supekar said.
What was different?
Using 678 brain scans of autistic children, the researchers developed an algorithm that could distinguish between boys and girls with 86% accuracy. When they checked the algorithm on the remaining 95 brain scans of autistic children, it retained the same accuracy in distinguishing boys from girls.
The scientists also tested the algorithm on 976 brain scans of typically developing boys and girls. The algorithm could not tell them apart, confirming that the sex differences discovered by the scientists were unique to autism.
Among autistic children, girls had different connectivity patterns than boys in several brain centers, including motor, linguistic, and visuospatial attention systems. Differences in a group of motor areas -; including primary motor cortex, supplementary motor area, parietal and lateral occipital cortex, and middle and superior temporal gyrus -; were the largest of the sexes. In autistic girls, differences in motor centers were related to the severity of their motor symptoms, meaning that girls whose brain patterns were most similar to those of autistic boys tended to have the most pronounced motor symptoms.
The researchers also identified language domains that differed between boys and girls with autism, and noted that previous studies had identified greater language impairment in boys.
“When you see that there are differences in brain regions that are linked to clinical symptoms of autism, it feels more real,” Supekar said.
Taken together, the findings should be used to guide future efforts to improve the diagnosis and treatment of girls, the researchers said.
“Our research advances the use of artificial intelligence-based techniques for precision psychiatry in autism,” Menon said.
“We may need to have different tests for females compared to males. The artificial intelligence algorithms we have developed can help improve the diagnosis of autism in girls,” Supekar said. At the treatment level, interventions for girls could be initiated earlier, he added.
Other co-authors of the Stanford Medicine study are data scientist Carlo de los Angeles; Principal Investigator Srikanth Ryali, PhD; and graduate student Kaidi Cao. Co-authors include members of the Stanford Maternal and Child Health Research Institute, Stanford Bio-X, Stanford Wu Tsai Neurosciences Institute, and Stanford Wu Tsai Human Performance Alliance, as well as the Stanford Institute for Human-Centered Artificial Intelligence.
The research was supported by the National Institutes of Health (grants AG072114, MH084164, and MH221069), the Brain & Behavior Research Foundation, a Stanford Innovator Award, and grants from the Stanford Maternal and Child Health Research Institutes, including the Initiatives Program transdisciplinary, the Taube Maternal and Child Health Research Fund and the Uytengsu-Hamilton Neuropsychiatry Research Program 22q11.
Supekar is a transdisciplinary researcher endowed with the Taube family for maternal and child health.
Supekar, K. et al. (2022) Deep learning identifies strong gender differences in functional brain organization and their dissociable links to clinical symptoms of autism. The British Journal of Psychiatry. doi.org/10.1192/bjp.2022.13.