Using AI, researchers find that brain organisation can differ between boys and girls with autism – because of gender disparities in generally male-focused research
Autism is a developmental disorder which can vary in spectrums of severity. It can affect children through development of social and communication insufficiencies, with symptoms such as demonstrating lessened interests or displays of repetitive behaviours.
In a study from the Stanford University School of Medicine, researchers aimed to analyse why autism symptoms can differ between girls and boys, potentially providing better diagnostics for girls.
Biased toward male patients, the original description of autism, published in 1943 by Leo Kanner, stated that boys were more likely to be diagnosed with autism. With male diagnoses four times higher in boys than in girls, most autism research has generally focused on, and consequentially benefitted males.
Now, researchers aim to recognise the differences autism can have on gender, using hundreds of brain scans with artificial intelligence techniques. They found that girls with brain patterns which were most like boys with autism tended to have the most pronounced motor symptoms.
Additionally, they identified that boys have greater language impairments.
Autism treatments are more effective during preschool years, during brain development
Published in The British Journal of Psychiatry, the study found that girls with autism often have fewer overt repetitive behaviours than boys, which could account for diagnostic delays.
The study analysed functional magnetic resonance imaging brain scans from 773 children with autism, with 637 boys and 136 girls. The small number of girls historically included in autism research has been a barrier to learning more about them, and still remains a barrier today as researchers note finding enough girls for the study was a struggle.
The study’s lead author, Kaustubh Supekar, PhD, said: “When a condition is described in a biased way, the diagnostic methods are biased. This study suggests we need to think differently.”
Senior author, Vinod Menon, added: “We detected significant differences between the brains of boys and girls with autism, and obtained individualized predictions of clinical symptoms in girls.
“We know that camouflaging of symptoms is a major challenge in the diagnosis of autism in girls, resulting in diagnostic and treatment delays.”
Lawrence Fung, MD, PhD, assistant professor of psychiatry and behavioural sciences, said: “Knowing that males and females don’t present the same way, both behaviourally and neurologically, is very compelling.
“If the treatments can be done at the right time, it makes a big, big difference: For instance, children on the autism spectrum receiving 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 up.
“If a child cannot articulate themselves well, they fall behind in many different areas. The consequences are really serious if they are not getting diagnoses early.”
The algorithm had 86% accuracy identifying boys and girls
For the prevalence of boys in the brain-scan databases, researchers set up a mathematical challenge where they used standard statistical methods used to find differences between groups, which required that the groups be reasonably equal in size.
The methods underlie machine-learning techniques in which algorithms can be trained to find patterns in very large and complex datasets.
Using 678 of the brain scans from children with autism, an algorithm was developed which could distinguish between boys and girls with 86% accuracy – verifying the algorithm on the remaining 95 brain scans from children with autism, it withheld the same accuracy at distinguishing boys from girls.
Additionally, they also tested the algorithm on 976 brain scans from typically developing boys and girls. However, the algorithm could not distinguish among them, confirming that the sex differences the scientists found were unique to autism.
What are the differences in boys and girls?
Girls with autism had different patterns of connectivity than boys did in several brain centres, such as seen with motor, language and visuospatial attention systems. Girl’s differences in motor canters were linked to the severity of their motor symptoms – which highlighted those girls whose brain patterns were most similar to boys with autism often had the most distinct motor symptoms.
Differences in a group of motor areas, including the primary motor cortex, supplementary motor area, parietal and lateral occipital cortex, and middle and superior temporal gyri, were seen to be the largest between the genders.
Supekar continued: “When I tried to identify differences (with traditional methods), the algorithm would tell me every brain is a male with autism. It was over-learning and not distinguishing between males and females with autism.
“When you see that there are differences in regions of the brain that are related to clinical symptoms of autism, this seems more real.”
Menon finalised: “Our research advances use of artificial intelligence-based techniques for precision psychiatry in autism.”
The researchers hope this study can be used to pilot future efforts to improve diagnosis and treatment for girls, lessening the diagnostic disparity.