| Summary: | 碩士 === 國立臺灣科技大學 === 醫學工程研究所 === 106 === Spatiotemporal parameters of gait are general gait parameters that can enable the simplest form of objective gait evaluation. These parameters tend to change in most locomotor disabilities and serve as guidelines for evaluating the walking ability of a subject. Wearable devices with sensors, such as accelerometers and gyroscopes, have been extensively used not only for gait analysis but also for fall detection and monitoring of physical activity and classification of posture and movement. In gait analysis, wearable technologies are mostly used to extract the spatiotemporal parameters of gait.
This study aimed to validate the application of an inertial measurement unit (IMU) sensor for calculating the spatiotemporal parameters of gait in healthy young, older adults, and stoke patients. Six healthy young adults, 4 older adults, and 10 stroke patients participated in this study; each wore an IMU sensor on the fifth lumbar vertebra (L5) while the sensor was synchronized with a three-dimensional motion capture system. Nineteen gait parameters, including duration, variability, and asymmetry, were calculated from initial contact and final contact event times. The statistical analysis result showed that all three groups has good to excellent agreement for four most commonly reported gait characteristic (stride-, step-, stance-,and swing-duration). Meanwhile, for asymmetry and variability, good agreement between two systems only found on subject with stroke group.
The second aim for this study is to classified the gait patterns of normal and stroke participants by using time- and frequency-domain features obtained from data provided by an Inertial Measurement Unit (IMU) sensor placed on the subject’s lower back (L5). Twenty-three participants were included and divided into two groups: healthy group (young and older adults) and stroke group. A feature selection method comprising statistical analysis and Signal-to-Noise Ratio (SNR) calculation was used to reduce the number of features. The features were used to train four Support Vector Machine (SVM) kernels, and the results were subsequently compared. The quadratic SVM kernel had the highest accuracy (93.46%), as evaluated through cross validation. Moreover, when different datasets were used on model testing, both the quadratic and cubic kernels showed the highest accuracy (96.55%). These results demonstrate the effectiveness of this study’s classification method in distinguishing between normal and stroke gait patterns.
|
|---|
