Nonparametric motion control in functional connectivity studies in children with autism spectrum disorder

Area of Science:

• Neuroscience
• Developmental Neuroscience
• Medical Imaging

Background:

• Autism spectrum disorder (ASD) is a neurodevelopmental condition impacting social interaction and communication.
• Resting-state functional magnetic resonance imaging (fMRI) is used to study brain functional connectivity in ASD.
• Head motion during fMRI scans introduces artifacts, complicating analysis and often leading to participant exclusion.

Purpose of the Study:

• To develop a novel method for analyzing functional connectivity in ASD that accounts for head motion without excluding participants.
• To quantify the difference in functional connectivity between autistic and non-ASD children while standardizing for motion.
• To introduce a nonparametric estimator, Motion Controlled (MoCo), for robust motion control in fMRI studies.

Main Methods:

• Proposed an estimand to standardize motion relative to low-motion scans.
• Introduced the Motion Controlled (MoCo) nonparametric estimator using an ensemble of machine learning methods.
• Applied the MoCo framework to fMRI data from 132 autistic and 245 non-ASD children.

Main Results:

• The MoCo estimator utilizes all participants, unlike traditional exclusion methods.
• MoCo significantly reduces motion artifacts compared to standard approaches without participant removal.
• The method demonstrates large-sample efficiency and robustness, accounting for potential selection biases.

Conclusions:

• The MoCo framework offers a more accurate and efficient way to analyze functional connectivity in ASD by effectively managing motion artifacts.
• This approach enhances data utilization and reduces bias, providing a more reliable estimation of brain connectivity differences in ASD.
• MoCo represents a significant advancement in neuroimaging analysis for neurodevelopmental disorders.