Aperiodic parameters of the fMRI power spectrum associate with preterm birth and neonatal age

Area of Science:

• Neuroscience
• Developmental Neuroscience
• Medical Imaging

Background:

• The perinatal period is crucial for infant brain development.
• Premature birth increases the risk of neurodevelopmental disorders.
• Previous studies show strong links between structural MRI and preterm birth, but weaker links with functional MRI (fMRI).

Purpose of the Study:

• To investigate the associations of the aperiodic component of the blood-oxygen-level-dependent (BOLD) signal power spectrum with preterm birth.
• To map these associations with post-menstrual age (PMA), postnatal age, and sex.
• To assess the utility of fMRI-derived parameters for predicting infant age using machine learning.

Main Methods:

• Utilized the task-free neonatal fMRI dataset from the Developing Human Connectome Project (dHCP).
• Analyzed the aperiodic component of the BOLD signal power spectrum from pre- and postcentral gyri.
• Employed machine learning regression to predict participant age based on aperiodic parameters from 90 cortical and subcortical regions.

Main Results:

• Identified significant differences in the aperiodic BOLD signal component between preterm and full-term infants.
• Found distinct associations of the aperiodic component with PMA, postnatal age, and sex.
• Achieved relatively high accuracy (test R² 0.20-0.41) in predicting participant age using machine learning regression on aperiodic parameters.

Conclusions:

• The aperiodic component of the BOLD signal in neonatal fMRI is sensitive to preterm birth and sex.
• fMRI-derived aperiodic parameters show potential for non-invasively assessing neurodevelopmental trajectories and predicting age in infants.
• This approach may offer new insights into brain development and inform early identification of neurodevelopmental conditions.