| The diagnosis of carpal tunnel syndrome (CTS) is based on a combination of characteristic symptoms and electrophysiologic abnormalities. However, the electrodiagnostic examination remains an expensive and time-consuming procedure. Therefore, more investigators have reported on their experiences on CTS diagnosis with ultrasound image. CTS is generally attributed to external compression of the median nerve. Thus, distinguishing the median nerve shape from speckle-obscured ultrasound image provides a potential for the diagnosis of CTS. To lower the speckle interference, we propose a novel compound imaging strategy to reveal the significant signatures of median nerve. We observed that patient¡¦s fingers motion in ultrasound imaging deforms nerve shape and produces tissue movement. Based on speckle decorrelation of image under different fingers motion conditions, speckle variations can be reduced by compounding technique. The primary assumption of compounding is that the ensemble average of a speckle image is the same as the incoherent average of the original object. In this study, the multi-level block matching and block sum pyramid integrated algorithms were adopted for speckle tracking and compounding between ultrasound images. The results indicate that the efficacy of speckle reduction is clear and the positions of the two median nerve boundaries are unchanged. In particular, the internal punctuated echoes of medial nerve become more distinct. |