当前位置: X-MOL 学术Med. Biol. Eng. Comput. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Classifying sitting, standing, and walking using plantar force data
Medical & Biological Engineering & Computing ( IF 2.6 ) Pub Date : 2021-01-08 , DOI: 10.1007/s11517-020-02297-4
Kohle J Merry 1 , Evan Macdonald 1 , Megan MacPherson 2 , Omar Aziz 1 , Edward Park 1 , Michael Ryan 3, 4 , Carolyn J Sparrey 1
Affiliation  

Prolonged static weight-bearing at work may increase the risk of developing plantar fasciitis (PF). However, to establish a causal relationship between weight-bearing and PF, a low-cost objective measure of workplace behaviors is needed. This proof-of-concept study assesses the classification accuracy and sensitivity of low-resolution plantar pressure measurements in distinguishing workplace postures. Plantar pressure was measured using an in-shoe measurement system in eight healthy participants while sitting, standing, and walking. Data was resampled to simulate on/off characteristics of 24 plantar force sensitive resistors. The top 10 sensors were evaluated using leave-one-out cross-validation with machine learning algorithms: support vector machines (SVMs), decision tree (DT), discriminant analysis (DA), and k-nearest neighbors (KNN). SVM and DT best classified sitting, standing, and walking. High classification accuracy was obtained with five sensors (98.6% and 99.1% accuracy, respectively) and even a single sensor (98.4% and 98.4%, respectively). The central forefoot and the medial and lateral midfoot were the most important classification sensor locations. On/off plantar pressure measurements in the midfoot and central forefoot can accurately classify workplace postures. These results provide the foundation for a low-cost objective tool to classify and quantify sedentary workplace postures.

Graphical abstract



中文翻译:

使用足底力数据对坐、站和步行进行分类

工作中长时间静态负重可能会增加患足底筋膜炎 (PF) 的风险。然而,要建立负重和 PF 之间的因果关系,需要对工作场所行为进行低成本的客观测量。这项概念验证研究评估了低分辨率足底压力测量在区分工作场所姿势方面的分类准确性和敏感性。使用鞋内测量系统测量八名健康参与者在坐、站和行走时的足底压力。重新采样数据以模拟 24 个足底力敏电阻器的开/关特性。使用留一法交叉验证和机器学习算法评估前 10 个传感器:支持向量机 (SVM)、决策树 (DT)、判别分析 (DA) 和 k-最近邻 (KNN)。SVM 和 DT 最好对坐、站和走进行分类。使用五个传感器(分别为 98.6% 和 99.1% 的准确度)甚至单个传感器(分别为 98.4% 和 98.4%)获得了高分类精度。中央前足和内侧和外侧中足是最重要的分类传感器位置。中足和前足中部的开/关足底压力测量可以准确地对工作场所姿势进行分类。这些结果为低成本的客观工具提供了基础,用于对久坐的工作场所姿势进行分类和量化。中央前足和内侧和外侧中足是最重要的分类传感器位置。中足和前足中部的开/关足底压力测量可以准确地对工作场所姿势进行分类。这些结果为低成本的客观工具提供了基础,用于对久坐的工作场所姿势进行分类和量化。中央前足和内侧和外侧中足是最重要的分类传感器位置。中足和前足中部的开/关足底压力测量可以准确地对工作场所姿势进行分类。这些结果为低成本的客观工具提供了基础,用于对久坐的工作场所姿势进行分类和量化。

图形概要

更新日期:2021-01-10
down
wechat
bug