Ultrasound Plane Wave Imaging as an Inverse Problem With Joint Sparse-Based Regularized Solution.

Area of Science:

• Medical Imaging
• Ultrasound Technology
• Signal Processing

Background:

• Traditional sparse regularization methods in ultrasound imaging face limitations in preserving speckle textures.
• Enhancing resolution, contrast, and texture fidelity in plane wave imaging is crucial for diagnostic accuracy.

Purpose of the Study:

• To develop a novel joint sparse regularization model to overcome limitations of existing methods.
• To improve resolution, contrast, and speckle texture preservation in plane wave ultrasound imaging.

Main Methods:

• A joint sparse regularization model combining ℓ1 (spatial sparsity) and ℓ2,1 (frequency domain joint sparsity) was proposed.
• The Alternating Direction Method of Multipliers (ADMM) algorithm was used to solve the inverse problem.
• Performance was evaluated on the PICMUS dataset and compared against Delay-and-Sum (DAS), ℓ1, and multi-plane wave compounding (75_PW) methods.

Main Results:

• The joint sparse method achieved superior axial and lateral resolutions (FWHM) compared to DAS and ℓ1 methods.
• High Contrast-to-Noise Ratio (CNR) values were obtained, passing the Kolmogorov-Smirnov (KS) test, indicating preserved speckle distribution.
• Reconstruction time was significantly reduced (5.3s) compared to ℓ1 and 75_PW methods, demonstrating improved computational efficiency.

Conclusions:

• The joint sparse regularization model effectively enhances ultrasound image resolution and contrast.
• Exploiting frequency-domain structure correlations preserves vital speckle textures.
• This method offers an efficient and robust solution for ultrasound inverse problem reconstruction.