Coordinate-Independent Ultrasound Assessment for Complex Tissue Kinematics: A Preliminary Study on Median Nerve Mobility.

Abstract

This paper presents an ultrasound strain-divergence imaging technique, which quantifies complex tissue kinematics without being constrained by imaging coordinates. An in vivo application of the technique was demonstrated on median nerve (MN) mobility. Transverse mobility of MN, a potential biomarker for evaluating MN entrapment in carpal tunnel syndrome (CTS), is featured with large variations across different active gestures, individuals, and operators, so it remains inconclusive in literature. Ultrasound data of the MN in the transverse view during active thumb and finger gestures at five wrist angles were acquired synchronously with electromyography and optical tracking of hand skeletons from 25 asymptomatic participants and 5 patients (7 CTS wrists). Two-dimensional mapping of both strain tensor ${\bm{E}}( {{{{\bm{E}}}_{{\bm{yy}}}},{\bm{\ }}{{{\bm{E}}}_{{\bm{xx}}}},{\bm{\ }}{{{\bm{E}}}_{{\bm{xy}}}}} )$ and motion divergence ${\bm{D}}$ was obtained for MN kinematics assessment. In the healthy group, statistical analysis revealed significant MN kinematic adaptation to different wrist postures by the regional median value of MN strains ${\tilde{\bm{E}}}$ (thumb motion: p<.001; finger motion: p<.001) and that of divergence ${\tilde{\bm{D}}}$ (thumb motion: p=.008; finger motion: p<.001) averaged from three repeated experiments. Thanks to its inherent coordinate-independence, ${\tilde{\bm{D}}}$ consistently showed local dilation/compression at volar-extended/flexed wrist positions. Comparison of ${\tilde{\bm{E}}}$ and ${\tilde{\bm{D}}}$ between healthy and CTS groups through linear regression underlined a decreased ability of the MN to adapt to different compressional levels, mimicked by different wrist angles, in CTS patients. The proposed ultrasound strain-divergence imaging is recommended for comprehensive kinematics assessment of biological tissues with an ill-defined geometry, complex motion patterns, or a large intra-/inter-subject variance.