Computational validation of optical adaptive depth steering for continuous-wave fNIRS

Area of Science:

• Neuroimaging
• Biomedical Optics
• Signal Processing

Background:

• Continuous-wave functional near-infrared spectroscopy (CW-fNIRS) relies on fixed optode placements.
• Inter-subject anatomical variability (skull thickness) and phenotypic factors (hair) cause significant bias in CW-fNIRS.
• Existing methods struggle to mitigate these biases effectively.

Purpose of the Study:

• To introduce the Optical Adaptive Depth Steering (O-ADS) framework for CW-fNIRS.
• To enhance brain sensitivity and signal recovery by overcoming anatomical and phenotypic limitations.
• To develop a software-defined solution for more inclusive cerebral monitoring.

Main Methods:

• Developed the O-ADS framework using a high-density multi-distance optode array (8-32 mm) and a linearly constrained minimum variance (LCMV) spatial beamformer.
• Synthesized "Virtual Optodes" to adapt to individual anatomy.
• Validated computationally using Monte Carlo simulations on cranial slabs and a 3D anatomical atlas (Colin27).

Main Results:

• O-ADS demonstrated up to a 39-fold improvement in relative brain sensitivity for thick-skull morphologies compared to static sensors.
• O-ADS achieved significantly higher mean signal recovery purity (38.3%) than standard Multi-Distance Regression (21.8%) under noise and simulated barriers.
• The framework showed high computational resilience to anatomical bias.

Conclusions:

• O-ADS offers a scalable solution for optical neuroimaging, reducing reliance on individual anatomical and phenotypic variance.
• This approach facilitates more inclusive cerebral monitoring, even under challenging physical constraints.
• O-ADS represents a significant advancement in mitigating biases in CW-fNIRS.