A multi-university research team, co-led by University of Virginia professor Gustavo K. Rohde, has developed a groundbreaking system that detects genetic markers of autism in brain images with an impressive accuracy of 89 to 95%.

This study emphasizes the strong genetic basis of autism, advocating for a genetics-first approach that could transform the understanding and treatment of the condition.

The transport-based morphometry (TBM) method utilized in this research allows for earlier diagnosis and personalized treatment of autism by focusing on genetic markers rather than waiting for behavioral cues.

The findings may help identify specific brain regions and mechanisms that can be targeted for therapy, potentially leading to more effective treatments for autism.

Published in the journal Science Advances on June 12, 2024, the research highlights a significant shift towards a genetics-first approach in understanding autism.

The TBM technique distinguishes normal brain structure variations from those linked to autism-related genetic variations known as copy number variations (CNVs).

The research team utilized data from the Simons Variation in Individuals Project, carefully matching control subjects for age, sex, handedness, and non-verbal IQ to ensure accurate comparisons.

Unlike traditional machine learning methods, TBM is based on mass transport principles, allowing for a more biologically relevant analysis of medical images.

This innovative approach overcomes limitations of conventional methods, which often fail to capture the biophysical connections inherent in the data.

The research received funding from several prominent organizations, including the National Science Foundation and the National Institutes of Health.

The research team included collaborators from the University of California San Francisco and Johns Hopkins University, with Shinjini Kundu, a former Ph.D. student of Rohde, serving as the first author.

The study's findings could revolutionize the diagnosis and treatment of autism, enabling earlier interventions based on genetic markers.