Tikhonov regularization-based extended Kalman filter technique for robust and accurate reconstruction in diffuse...

Area of Science:

• Biomedical Optics
• Medical Imaging
• Computational Imaging

Background:

• Diffuse Optical Tomography (DOT) is a non-invasive imaging technique using near-infrared light.
• Conventional DOT methods lack robustness to measurement errors, impacting image quality.
• Deterministic DOT inversion methods often ignore measurement errors, leading to poor noise handling.

Purpose of the Study:

• To improve DOT image reconstruction quality and noise robustness.
• To develop an Extended Kalman Filter (EKF)-based DOT algorithm incorporating a priori information and measurement errors.
• To introduce Tikhonov regularization into the EKF for enhanced stability in ill-conditioned tomographic problems.

Main Methods:

• Development of an Extended Kalman Filter (EKF)-based DOT reconstruction algorithm.
• Integration of Tikhonov regularization with the EKF to address ill-posedness.
• Validation through numerical simulations and phantom experiments, comparing with Algebraic Reconstruction Technique and Levenberg-Marquardt.

Main Results:

• EKF-based algorithms accurately estimate target location and size.
• Significant improvements in imaging accuracy and noise robustness compared to conventional methods.
• Tikhonov-regularized EKF demonstrated optimal parameter estimation, particularly in low contrast and high noise scenarios.

Conclusions:

• EKF-based algorithms offer superior performance for DOT imaging.
• Tikhonov regularization enhances the stability and accuracy of EKF for DOT.
• These advanced methods improve the reliability of DOT for probing tissue optical properties.