Assessment of median nerve mobility by ultrasound dynamic imaging in carpal tunnel syndrome diagnosis

Abstract

Carpal tunnel syndrome (CTS) is a common entrapment neuropathy. Nerve conduction studies (NCS) have been used as a standard for CTS diagnosis. Complementing NCS, ultrasound imaging provides anatomic information on pathologic changes of the median nerve, such as the reduced median nerve mobility. Motion of median nerve is dependent on mechanical characteristics, and body movements. The purpose of this study was therefore to measure transverse sliding patterns of the median nerve during fingers flexion and extension in ultrasound B-mode images for distinguishing healthy from CTS subjects, and to investigate any correlation between NCS severity and median nerve motion. Transverse ultrasound images were acquired from 19 normal, 15 mild, and 10 severe CTS subjects confirmed by NCS. In two-second acquisition, their fingers were initially in natural position; the median nerve was then moved toward the ulnar side and radius side in fingers flexion and extension, respectively. The displacements of the median nerve were calculated by the multilevel block-matching pyramid algorithm and averaged. All the average displacements at different acquisition times were then accumulated to obtain cumulative displacements, which were curve-fitted by polynomial function. To differentiate the normal from CTS cases, the R-squared, curvature, and amplitude of the fitted curves were computed, to evaluate the goodness, variation, and maximum value of the fit, respectively. Compared to the CTS patients, the normal subjects had higher R-square, curvature, and amplitude estimates. The three parameters were then inputted to a fuzzy c-means algorithm to classify normal cases and CTS ones. The diagnostic efficiency had an accuracy of 93.2%, a specificity of 100%, and a sensitivity of 88%. Further study includes measuring mechanical strain and stress at different neural sites to provide elasticity of the median nerve.