碩士 / 國立成功大學 / 電腦與通信工程研究所 / 103 / The purpose of the thesis is use inertial sensors to collect the gait signal and reconstruct the participants’ gait trajectory. Then, we can classify the participants are Alzheimer disease patients or not. We designed a gait experiment including single-task and dual-task to analysis the participants’ dynamic balance ability. At first, participants put the inertial sensor devices on their right instep. Then, the inertial sensor devices collected the participants’ signal during the experiment. After signal preprocessing, we can use the signal to reconstruction three-dimension trajectory. After this, we should verify the trajectory’s rationality with the motion capture system. We use the correlation analysis to get the correlation coefficient between the inertial-sensor based trajectory and the motion capture system trajectory. We found that the two trajectory are highly correlated. After getting the trajectory, we can get the gait features including stride height, stride number, and stride length at the same time. Then, we use support vector machine-based classifier to separate Alzheimer’s disease patients and healthy participants. We can get the results after the classification. The optimal accuracy was closed to 85 percent, and the optimal specificity was closed to 83 percent.
| Identifer | oai:union.ndltd.org:TW/103NCKU5652060 |
| Date | January 2015 |
| Creators | Yu-TingHung, 洪御庭 |
| Contributors | Pau-Choo Chung, Ming-Chyi Pai, 詹寶珠, 白明奇 |
| Source Sets | National Digital Library of Theses and Dissertations in Taiwan |
| Language | zh-TW |
| Detected Language | English |
| Type | 學位論文 ; thesis |
| Format | 57 |
Page generated in 0.0168 seconds