Dynamic causal modelling on infant fNIRS data: A validation study on a simultaneously recorded fNIRS-fMRI dataset

Area of Science:

• Neuroscience
• Developmental Neuroscience
• Brain Connectivity

Background:

• Tracking infant brain connectivity is crucial for developmental neuroscience.
• Functional near-infrared spectroscopy (fNIRS) is ideal for infant brain studies due to its portability and safety.
• Existing fNIRS data analysis methods are less developed than those for fMRI, particularly for connectivity.

Purpose of the Study:

• To provide a proof-of-principle for applying Dynamic Causal Modelling (DCM) to infant fNIRS data.
• To demonstrate the robustness of DCM for infant fNIRS using simultaneous fMRI-fNIRS recordings.
• To establish a data analysis pipeline for future infant effective connectivity research.

Main Methods:

• Simultaneous fMRI and fNIRS recording from a 6-month-old infant during auditory stimulation.
• Preprocessing of fMRI and fNIRS data using SPM and general linear model.
• Adapting DCM for infant fNIRS, addressing challenges in structural image import, spatial registration, segmentation, meshing, and optical sensitivity estimation.

Main Results:

• High correspondence in variational Free Energy (F), Bayesian Model Selection (BMS), and Bayesian Model Average (BMA) between fMRI and fNIRS data.
• Demonstrated high reliability of DCM when applied to infant fNIRS data.
• Successfully adapted DCM for infant fNIRS, overcoming key technical challenges.

Conclusions:

• Dynamic Causal Modelling (DCM) is a reliable method for analyzing effective connectivity in infant fNIRS data.
• This study provides a validated pipeline and guidance for applying DCM to infant fNIRS.
• Opens new research avenues for understanding brain connectivity in infancy.