Vector correlation technique for pixel-wise detection of collagen fiber realignment during injurious tensile loading

Excessive soft tissue loading can produce adverse structural and physiological changes in the absence of any visible tissue rupture. However, image-based analysis techniques to assess microstructural changes during loading without any visible rupture remain undeveloped. Quantitative polarized light imaging (QPLI) can generate spatial maps of collagen fiber alignment during loading with high temporal resolution and can provide a useful technique to measure microstructural responses. While collagen fibers normally realign in the direction that tissue is loaded, rapid, atypical fiber realignment during loading may be associated with the response of a local collagenous network to fiber failure. A vector correlation technique was developed to detect this atypical fiber realignment using QPLI and mechanical data collected from human facet capsular ligaments (n=16) loaded until visible rupture. Initial detection of anomalous realignment coincided with a measurable decrease in the tissue stiffness in every specimen and occurred at significantly lower strains than those at visible rupture (p<0.004), suggesting this technique may be sensitive to a loss of microstructural integrity. The spatial location of anomalous realignment was significantly associated with regions where visible rupture developed (p<0.001). This analysis technique provides a foundation to identify regional differences in soft tissue injury tolerances and relevant mechanical thresholds.