Singular Spectrum Analysis (SSA) Based Tremor Supression In Micro-Surgery

Abstract

Tremor as a result of unstable nature of human organ is one of the main problems in micro surgery due to its effects in generating undesirable surgical outcomes. The unwanted movement impedes microsurgery and lead to poor performance. Various techniques were proposed to improve the effect of tremor to achieve high level of precision. In this work, SSA algorithm is developed to suppress the physiological tremor present in the hands of surgeon. The technique decomposes the time domain signal into different singular spectrum (interpretable components) domain via the singular value decomposition (SVD), in which primary signal such as trends and oscillatory components can be recognized, isolated from the tremor and regrouped in a linear fashion, and finally time domain signals reconstructed. The developed algorithm promises potential for tremor suppression in realtime. By choosing a group of specific decomposed signals based on their eigenvalues and spectral range, both the primary and tremor motions can be reconstructed accurately. The algorithm shows the tremor signal can be estimated from the total motion with an accuracy greater than 89%.