Ultrasound imaging settings can optimize spatial frequency analysis and pathology discrimination in supraspinatus tendon

Citation Author(s):
Oscar
Vila-Dieguez
Submitted by:
Oscar Vila-Dieguez
Last updated:
Mon, 07/15/2024 - 11:14
DOI:
10.21227/k119-w370
License:
157 Views
Categories:
Keywords:
0
0 ratings - Please login to submit your rating.

Abstract 

Ultrasound (US) provides non-invasive visualization of tissue morphology for musculoskeletal disorders. Spatial Frequency Analysis (SFA) of US images quantitatively characterizes tissue morphology, and has shown the ability to distinguish healthy from pathological tendons. However, the impact of US machine settings on SFA for tendon pathology remains underexplored. Methods: Five participants with unilateral supraspinatus tendon partial tears were imaged bilaterally to examine how variations in US settings (frequency, dynamic range, gain) influence SFA parameters. Tendons were scanned at 9 and 12 MHz with all possible dynamic range and gain combinations, resulting in 1680 images processed via SFA. A linear mixed-effects model analyzed the impact of US settings and tendon conditions on four SFA parameters. Results: The SFA peak spatial frequency radius (PSFR) significantly differed between healthy and pathological tendons, regardless of machine settings (Coef. = 0.55, SE = 0.04, p < 0.001). The 12 MHz frequency notably enhanced PSFR’s discriminatory power, showing higher values in healthy tendons (mean PSFR = 1.34 mm⁻¹) and lower in partial tears (mean PSFR = 1.09 mm⁻¹) compared to 9 MHz (healthy = 1.22 mm⁻¹, partial tear = 1.04 mm⁻¹). P6, Q6, and Amax parameters were affected by machine settings. P6 and Q6 differentiated tendon conditions but required consistent dynamic range settings. Conclusion: Optimal US settings enhance SFA's ability to discriminate between healthy and pathological supraspinatus tendons, with PSFR as the most robust parameter. 12 MHz significantly improves PSFR’s discriminatory capability. Consistency in US parameters is crucial for reliable SFA results, particularly for P6 and Q6.

Instructions: 

Image per row. Spatial frequency analysis parameter per column