Liu Yixin, He Xiaohai, Wang Renjie, Teng Qizhi, Hu Rui, Qing Linbo, Wang Zhengyong, He Xuan, Yin Biao, Mou Yi, Du Yanping, Li Xinyi, Wang Hui, Liu Xiaolei, Zhou Lixing, Deng Linghui, Xu Ziqi, Xiao Chun, Ge Meiling, Sun Xuelian, Jiang Junshan, Chen Jiaoyang, Lin Xinyi, Xia Ling, Gong Haoran, Yu Haopeng, Dong Birong
National Clinical Research Center for Geriatrics, West China Hospital, Sichuan University, Chengdu, China.
Geriatric Health Care and Medical Research Center, Sichuan University, Chengdu, China.
Front Aging Neurosci. 2021 Nov 16;13:757823. doi: 10.3389/fnagi.2021.757823. eCollection 2021.
Frail older adults have an increased risk of adverse health outcomes and premature death. They also exhibit altered gait characteristics in comparison with healthy individuals. In this study, we created a Fried's frailty phenotype (FFP) labelled casual walking video set of older adults based on the West China Health and Aging Trend study. A series of hyperparameters in machine vision models were evaluated for body key point extraction (AlphaPose), silhouette segmentation (Pose2Seg, DPose2Seg, and Mask R-CNN), gait feature extraction (Gaitset, LGaitset, and DGaitset), and feature classification (AlexNet and VGG16), and were highly optimised during analysis of gait sequences of the current dataset. The area under the curve (AUC) of the receiver operating characteristic (ROC) at the physical frailty state identification task for AlexNet was 0.851 (0.827-0.8747) and 0.901 (0.878-0.920) in macro and micro, respectively, and was 0.855 (0.834-0.877) and 0.905 (0.886-0.925) for VGG16 in macro and micro, respectively. Furthermore, this study presents the machine vision method equipped with better predictive performance globally than age and grip strength, as well as than 4-m-walking-time in healthy and pre-frailty classifying. The gait analysis method in this article is unreported and provides promising original tool for frailty and pre-frailty screening with the characteristics of convenience, objectivity, rapidity, and non-contact. These methods can be extended to any gait-related disease identification processes, as well as in-home health monitoring.
体弱的老年人出现不良健康后果和过早死亡的风险增加。与健康个体相比，他们的步态特征也有所改变。在本研究中，我们基于华西健康与衰老趋势研究创建了一组标注有老年人日常步行视频的弗里德虚弱表型（FFP）。对机器视觉模型中的一系列超参数进行了评估，用于身体关键点提取（AlphaPose）、轮廓分割（Pose2Seg、DPose2Seg和Mask R-CNN）、步态特征提取（Gaitset、LGaitset和DGaitset）以及特征分类（AlexNet和VGG16），并在分析当前数据集的步态序列时进行了高度优化。在身体虚弱状态识别任务中，AlexNet的宏观和微观接收者操作特征（ROC）曲线下面积（AUC）分别为0.851（0.827 - 0.8747）和0.901（0.878 - 0.920），VGG16的宏观和微观AUC分别为0.855（0.834 - 0.877）和0.905（0.886 - 0.925）。此外，本研究提出的机器视觉方法在整体预测性能上优于年龄和握力，在健康和脆弱前期分类方面也优于4分钟步行时间。本文的步态分析方法未见报道，为虚弱和脆弱前期筛查提供了一种具有便利性、客观性、快速性和非接触性特点的有前景的原创工具。这些方法可扩展到任何与步态相关的疾病识别过程以及家庭健康监测。
