The role of head circumference and cerebral volumes to phenotype male adults with autism spectrum disorder. Issue 3 (3rd February 2022)

• Record Type: Journal Article
• Authors: Denier, Niklaus; Steinberg, Gerrit; van Elst, Ludger Tebartz; Bracht, Tobias
• Is Part Of: Brain and behavior. Volume 12:Issue 3(2022)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-02-03
• Subjects: autism spectrum disorder -- biomarker -- classification -- head circumference -- machine learning -- phenotype; Neurology -- Periodicals; Neurosciences -- Periodicals; Psychology -- Periodicals; Psychiatry -- Periodicals; 616.8005
• DOI: 10.1002/brb3.2460
• Languages: English
• ISSNs: 2162-3279

Abstract

Abstract: Background: Autism spectrum disorder (ASD) has been repeatedly associated with enlargements of head circumference in children with ASD. However, it is unclear if these enlargements persist into adulthood. This is the first study to investigate head circumference in a large sample of adults with ASD. Methods: We apply a fully automated magnetic resonance imaging (MRI) based measurement approach to compute head circumference by combining 3D and 2D image processing. Head circumference was compared between male adults with ASD ( n = 120) and healthy male controls ( n = 136), from the Autism Brain Imaging Data Exchange (ABIDE) database. To explain which brain alterations drive our results, secondary analyses were performed for 10 additional morphological brain metrics. Results: ASD subjects showed an increase in head circumference ( p = .0018). In addition, ASD patients had increased ventricular surface area (SA) ( p = .0013). Intracranial volume, subarachnoidal cerebrospinal fluid (CSF) volume, and gray matter volume explained 50% of head circumference variance. Using a linear support vector machine, we gained an ASD classification accuracy of 73% (sensitivity 92%, specificity 68%) using head circumference and brain‐morphological metrics as input features. Head circumference, ventricular SA, ventricular CSF volume, and ventricular asymmetry index contributed to 85% of feature weighting relevant for classification. Conclusion: Our results suggest that head circumference increases in males with ASD persist into adulthood. Results may be driven by morphological alterations of ventricular CSF. The presented approach for an automated head circumference measurement allows for the retrospective investigation of large MRI datasets in neuropsychiatric disorders. Abstract : We investigated by a novel and fully automated MRI‐based measurement approach head circumference and additional brain morphological metrics in a large publicly available dataset of an adult male with autism spectrum disorder (ASD) and healthy controls. We found increased head circumference in ASD, mostly explained by alterations in cerebrospinal fluid volume.