Accelerated image reconstruction using extrapolated Tikhonov filtering for photoacoustic tomography.

Area of Science:

• Medical Imaging
• Biomedical Engineering
• Computational Science

Background:

• Photoacoustic tomography (PAT) reconstruction commonly uses Tikhonov regularization.
• Automated regularization parameter selection is computationally intensive.
• Standard Tikhonov methods can result in loss of image sharpness.

Purpose of the Study:

• To develop simple and computationally efficient extrapolated Tikhonov filtering methods for PAT.
• To improve the quality and speed of image reconstruction in PAT.

Main Methods:

• Proposed an extrapolation method to estimate the solution at zero regularization.
• Applied the method to three variants of Tikhonov filtering (Lanczos, traditional, exponential).
• Compared the extrapolated methods against the standard error estimate technique.

Main Results:

• Demonstrated effectiveness using four numerical and two experimental phantoms.
• Achieved at least four times faster computation compared to standard methods.
• Showed up to 2.6 times improvement in standard figures of merit.

Conclusions:

• Extrapolated Tikhonov filtering overcomes challenges in regularization parameter selection.
• The methods reconstruct high-quality PAT images efficiently.
• The computational efficiency makes these methods suitable for real-time PAT applications.