Ultrasound Displacement Tracking Techniques for Post-Stroke Myofascial Shear Strain Quantification.

Area of Science:

• Musculoskeletal imaging
• Biomarker development
• Ultrasound elastography

Background:

• Ultrasound shear strain shows potential as a biomarker for myofascial dysfunction.
• Displacement tracking techniques can affect shear strain estimation accuracy.
• Clinical conclusions on myofascial pain may be influenced by tracking method variability.

Purpose of the Study:

• To assess the reliability of four ultrasound displacement estimation algorithms.
• To test the hypothesis that paretic shoulder muscles exhibit lower shear strain than non-paretic muscles in stroke patients with myofascial pain.

Main Methods:

• Evaluated four algorithms: Search, OVERWIND-Search, SOUL-Search, and L1-SOUL-Search.
• Algorithms were validated using simulations and in vivo data from ten stroke patients.
• Methods involved optimizing cost functions with L1- or L2-norm regularization for displacement estimation.

Main Results:

• SOUL-Search and L1-SOUL-Search demonstrated the highest accuracy and reliability in estimating shear strain.
• L1-SOUL-Search provided superior displacement tracking, yielding smooth gradients and distinct muscle layer edges.
• Results supported the clinical hypothesis of reduced shear strain in paretic shoulder muscles.

Conclusions:

• L1-SOUL-Search is the most suitable algorithm for investigating myofascial pain and dysfunction using ultrasound shear strain.
• While computationally intensive, L1-SOUL-Search's runtime may be improved with deep learning.
• This research lays the groundwork for developing reliable ultrasound-based biomarkers for musculoskeletal health.