Statistical gait analysis of difference between healthy humans and patients with injuries to the cervical spine

• Main Authors: Guan-Sheng Huang, 黃冠勝
• Other Authors: Chao-Cheng Wu
• Format: Others
• Language: en_US
• Published: 2013
• Online Access: http://ndltd.ncl.edu.tw/handle/85383902123299063813

碩士 === 國立臺北科技大學 === 電機工程系研究所 === 101 === Gait has been one of important indicators for healthcare systems. In neurosurgery, gait can be regarded as the coordinated action of nerves and muscles. Unfortunately, the current diagnosis is mainly based on the subjective judgment of doctor. It is difficult to establish long-term tracking and detect the subtle changes which are not visible and sensible by human beings. Gait analysis is a popular technique to assess and measure the ability to action of human from a macro point of view. It is important to build an objective and quantifying gait analysis. Traditional methods were often confined in the laboratory environment and record the actions of human gait by the sophisticated instrument. These methods are difficult to reflect actions in daily life. In order to avoid these constraints, this thesis exploited a smart phone with tri-axial accelerometer as the data source, and statistically analyzed signals collected from normal human and patients with injuries to the cervical spine. The results could offer additional diagnostic information for doctors and a recovery indicator for patients. This thesis first recognized seven stages of each gait cycle collected from a tri-axial accelerometer and then applied three popular signal analysis algorithms, Fourier Transform (FT), Short-Time Fourier Transform (STFT), Hilbert-Huang Transform (HHT). The statistical indicators would be built up for quantitative gait analysis. The contribution of this thesis aimed to provide important gait information which is difficult to be sensed by doctors and a second opinion for medical diagnosis. In addition, it could also be a recovery indicator for patients.