Wu Yunfeng, Ng Sin Chun
Department of Communication Engineering, School of Information Science and Technology, Xiamen University, 422 Si Ming South Road, Fujian, 361005, China.
Annu Int Conf IEEE Eng Med Biol Soc. 2010;2010:1304-7. doi: 10.1109/IEMBS.2010.5626398.
Amyotrophic lateral sclerosis (ALS) is a type of neurological disease due to the degeneration of motor neurons. During the course of such a progressive disease, it would be difficult for ALS patients to regulate normal locomotion, so that the gait stability becomes perturbed. This paper presents a pilot statistical study on the gait cadence (or stride interval) in ALS, based on the statistical analysis method. The probability density functions (PDFs) of stride interval were first estimated with the nonparametric Parzen-window method. We computed the mean of the left-foot stride interval and the modified Kullback-Leibler divergence (MKLD) from the PDFs estimated. The analysis results suggested that both of these two statistical parameters were significantly altered in ALS, and the least-squares support vector machine (LS-SVM) may effectively distinguish the stride patterns between the ALS patients and healthy controls, with an accurate rate of 82.8% and an area of 0.87 under the receiver operating characteristic curve.
肌萎缩侧索硬化症（ALS）是一种因运动神经元退化而导致的神经疾病。在这种进行性疾病的病程中，ALS患者很难调节正常的运动，从而使步态稳定性受到干扰。本文基于统计分析方法，对ALS患者的步态节奏（或步幅间隔）进行了一项初步统计研究。首先采用非参数Parzen窗方法估计步幅间隔的概率密度函数（PDF）。我们计算了左脚步幅间隔的均值以及根据估计的PDF得出的修正库尔贝克-莱布勒散度（MKLD）。分析结果表明，这两个统计参数在ALS患者中均有显著变化，并且最小二乘支持向量机（LS-SVM）能够有效区分ALS患者和健康对照者的步幅模式，准确率为82.8%，在受试者工作特征曲线下的面积为0.87。
