Retrieval procedure for time-resolved near-infrared tissue spectroscopy based on the optimal estimation method

Area of Science:

• Biomedical Optics
• Spectroscopy
• Computational Modeling

Background:

• Time-resolved near-infrared tissue spectroscopy (TR-NIRS) is crucial for non-invasive tissue analysis.
• Retrieving accurate optical properties from TR-NIRS data presents an ill-posed inverse problem.
• Standard nonlinear least-squares methods can be limited in precision and parameter recovery.

Purpose of the Study:

• To develop and validate an improved retrieval procedure for TR-NIRS data.
• To enhance the accuracy and precision of retrieved tissue optical properties.
• To address the ill-posed nature of the inverse problem in NIRS.

Main Methods:

• Utilized the optimal estimation method for parameter retrieval.
• Incorporated a priori information for both target and forward model parameters.
• Validated the method using simulated TR-NIRS experiments based on diffusion and radiative transfer equations.

Main Results:

• The optimal estimation method demonstrated a smaller difference between retrieved and true parameter values.
• Incorporating a priori information led to reduced error bars on retrieved parameters.
• Accurate estimation of forward model parameter errors improved the retrieval of target parameters.

Conclusions:

• Optimal estimation offers a robust approach for enhancing TR-NIRS data analysis.
• The method effectively mitigates challenges associated with ill-posed inverse problems.
• This technique holds potential for more precise non-invasive tissue characterization.