Image quality analysis of high-density diffuse optical tomography incorporating a subject-specific head model

Area of Science:

• Biomedical Engineering
• Neuroimaging
• Optical Physics

Background:

• Traditional near-infrared spectroscopy has limitations in brain imaging resolution and accuracy.
• High-density diffuse optical tomography (HD-DOT) offers improved performance for functional brain imaging.

Purpose of the Study:

• To comprehensively evaluate the image quality of HD-DOT for visual cortex mapping.
• To assess the impact of anatomical constraints on HD-DOT reconstruction accuracy and depth.

Main Methods:

• A simulation study using an MRI-derived anatomical head model.
• HD-DOT model with multiple source-detector separations and continuous-wave data with added noise.
• Quantification of localization error and localized volume at half maximum (LVHM).
• Systematic analysis of whole-brain tissue spatial constraints in image reconstruction.

Main Results:

• HD-DOT achieved localization error <10 mm and LVHM <1000 mm³ up to 13 mm below the scalp.
• Utilizing a whole-brain spatial constraint improved imaging depth to 18 mm.
• Anatomical modeling enhances the realism of the physical and inverse problem solutions.

Conclusions:

• HD-DOT provides high-quality imaging for visual cortex mapping.
• Anatomical constraints significantly improve the depth and accuracy of HD-DOT reconstructions.
• This simulation study validates HD-DOT's potential for detailed human brain imaging.