Three dimensional photoacoustic tomography in Bayesian framework.

Area of Science:

• Medical Imaging
• Biomedical Engineering
• Computational Science

Background:

• Photoacoustic tomography (PAT) reconstructs initial pressure distributions from ultrasound waves.
• A recent Bayesian approach with uncertainty quantification was limited to 2D simulations.
• Extending PAT reconstruction to 3D is crucial for clinical applications.

Purpose of the Study:

• To extend the Bayesian approach for photoacoustic image reconstruction to three spatial dimensions.
• To incorporate experimental data alongside numerical simulations.
• To assess the accuracy and reliability of the 3D reconstruction.

Main Methods:

• Utilized a Bayesian framework for image reconstruction and uncertainty quantification.
• Extended the method to three spatial dimensions.
• Employed iterative matrix-free computations with a biconjugate gradient stabilized method.
• Incorporated experimental data into the reconstruction process.

Main Results:

• The Bayesian approach successfully reconstructed initial pressure distributions in 3D.
• Accurate estimates were achieved even in realistic measurement geometries.
• The method provided reliable assessment of the uncertainty in the reconstructed images.
• Validation was performed using both numerical simulations and experimental data.

Conclusions:

• The 3D Bayesian approach is effective for photoacoustic image reconstruction.
• This method allows for accurate initial pressure distribution recovery in complex scenarios.
• Uncertainty quantification enhances the reliability and interpretability of PAT reconstructions.