Combining Total Variation Regularization with Window-Based Time Delay Estimation in Ultrasound Elastography.

Area of Science:

• Medical Imaging
• Biomedical Engineering
• Signal Processing

Background:

• Accurate displacement estimation is crucial for ultrasound elastography (USE).
• Existing methods for displacement estimation in USE face challenges with noise and ill-posed problems.
• Current approaches include window-based assumptions or smoothness regularization.

Purpose of the Study:

• To introduce a novel, robust method for displacement estimation in free-hand palpation ultrasound elastography.
• To incorporate L1 norm (total variation) regularization for improved sharpness and robustness.
• To evaluate the performance of the proposed method against existing techniques.

Main Methods:

• Developed tOtal Variation Regularization and WINDow-based time delay estimation (OVERWIND) method.
• Combined window-based constant displacement assumption with L1 norm regularization.
• Utilized an iterative optimization approach for the L1 norm cost function.

Main Results:

• OVERWIND demonstrates robustness to signal decorrelation.
• The method generates sharp displacement and strain maps for simulated, phantom, and in-vivo data.
• Significant improvements in strain contrast-to-noise ratio (CNR) were observed: 27.26% (simulation), 144.05% (phantom), and 49.90% (in-vivo) compared to the GLUE method.

Conclusions:

• OVERWIND offers a more robust and accurate approach to displacement and strain estimation in ultrasound elastography.
• The use of L1 norm regularization enhances the sharpness of displacement estimates.
• The proposed method shows significant potential for improving quantitative assessments in medical ultrasound imaging.