Model based reconstruction for simultaneously imaging cerebral blood flow and De-oxygen hemoglobin distribution

Area of Science:

• Biomedical Optics
• Neuroimaging
• Physiological Measurement

Background:

• Laser speckle imaging (LSI) and optical intrinsic signal (OIS) imaging provide high-resolution data on cerebral blood flow (CBF) and deoxygenated hemoglobin (dHb).
• Cortex curvature and non-uniform illumination introduce artifacts, distorting CBF and dHb measurements.
• Accurate quantification of these parameters is crucial for understanding brain function and disease.

Purpose of the Study:

• To develop a method for extracting accurate CBF and dHb images from raw speckle images.
• To address and correct for image distortions caused by physiological and optical factors.
• To improve the reliability of LSI and OIS in neuroimaging applications.

Main Methods:

• Extraction of both CBF and dHb images directly from raw speckle images.
• Estimation of a parametric model to account for image inhomogeneity.
• Application of adaptive window median-mean filtering and curve fitting techniques for reconstruction.

Main Results:

• Successfully reconstructed corrected CBF and dHb images.
• Demonstrated the ability to mitigate distortions from cortex curvature and illumination variations.
• Provided a more accurate representation of hemodynamic parameters in the brain.

Conclusions:

• The proposed method effectively extracts and reconstructs accurate CBF and dHb images from raw speckle data.
• This technique offers a significant improvement for quantitative analysis in LSI and OIS neuroimaging.
• The findings enhance the utility of optical imaging for studying brain physiology.