IEEE Trans Neural Syst Rehabil Eng. 2017 Nov;25(11):2018-2025. doi: 10.1109/TNSRE.2017.2699124. Epub 2017 Apr 27.
Frailty is characterized by a loss of functionality and is expected to affect 9.9% of people aged 65 and over. Here, current frailty classification is compared with a collection of selected kinematic parameters. A total of 718 elderly subjects (319 males and 399 females; age: 75.4 ± 6.1 years), volunteered to participate in this study and were classified according to Fried's criteria. Both the 30-s chair stand test (CST) and the 3-m walking test were performed and a set of kinematic parameters were obtained from a single inertial unit. A decision tree analysis was used to: 1) identify the most relevant frailty-related parameters and 2) compare validity of this classification. We found that a selected set of parameters from the 30-s CST (i.e., range of movement, acceleration, and power) were better at identifying frailty status than both the actual outcome of the test (i.e., cycles' number) and the normally used criteria (i.e., gait speed). For the pre-frail status, AUC improves from 0.531 using the actual test outcome and 0.516 with gait speed to 0.938 with the kinematic parameters criteria. In practice, this could improve the presyndrome identification and perform the appropriate actions to postpone the progression into the frail status.
虚弱的特点是功能丧失，预计 65 岁及以上的人群中有 9.9%会受到影响。在这里，当前的虚弱分类与一系列选定的运动学参数进行了比较。共有 718 名老年受试者（319 名男性和 399 名女性；年龄：75.4 ± 6.1 岁）自愿参加了这项研究，并根据 Fried 的标准进行了分类。所有受试者都进行了 30 秒椅子站立测试（CST）和 3 米步行测试，并从单个惯性单元中获得了一组运动学参数。决策树分析用于：1）确定与虚弱相关的最相关参数；2）比较这种分类的有效性。我们发现，30 秒 CST 的一组选定参数（即运动范围、加速度和功率）比实际测试结果（即循环次数）和常用标准（即步态速度）更能识别虚弱状态。对于前虚弱状态，使用实际测试结果和步态速度的 AUC 分别从 0.531 和 0.516 提高到运动学参数标准的 0.938。实际上，这可以改善预综合征的识别，并采取适当的行动来推迟进展到虚弱状态。
