Research on Prediction Model of Physical Activity Energy Expenditure with Wearable Sensors

doi:10.3778/j.issn.1673-9418.2104086

Journal of Frontiers of Computer Science and Technology ›› 2022, Vol. 16 ›› Issue (12): 2832-2840.DOI: 10.3778/j.issn.1673-9418.2104086

• Artificial Intelligence • Previous Articles Next Articles

Research on Prediction Model of Physical Activity Energy Expenditure with Wearable Sensors

WANG Lin¹, SUN Qian¹, MA Xiaona¹, GAO Yongyan², LIU Yi¹, MA Hongwei¹, YANG Dongqiang¹^,⁺()

1. College of Computer Science and Technology, Shandong Jianzhu University, Jinan 250101, China
2. Department of Sports, Shandong Jianzhu University, Jinan 250101, China

Received:2021-04-13 Revised:2021-06-01 Online:2022-12-01 Published:2021-06-08
About author:WANG Lin, born in 1975, M.S. candidate, lecturer. Her research interests include machine learning, intelligent city, etc.
SUN Qian, born in 1983, M.S. candidate, lecturer. Her research interest is intelligent city.
MA Xiaona, born in 1977, M.S. candidate, associate professor. Her research interests include software testing, machine learning, etc.
GAO Yongyan, born in 1980, M.S. candidate, associate professor. Her research interests include physical education teaching and training, physical culture, etc.
LIU Yi, born in 1978, M.S. candidate, lecturer. His research interests include machine vision, artificial intelligence, etc.
MA Hongwei, born in 1969, Ph.D. candidate, professor, M.S. supervisor. His research interests include computer system architecture, computer network, machine learning, etc.
YANG Dongqiang, born in 1970, Ph.D. candidate, associate professor, M.S. supervisor. His research interests include natural language processing, artificial intelligence, etc.
Supported by:
Natural Science Foundation of Shandong Province(ZR2018MF012);Natural Science Foundation of Shandong Province(ZR2020QF029);Ph.D. Foundation of Shandong Jianzhu University(XNBS1811)

可穿戴传感器的人体活动能量预测模型研究

王琳¹, 孙倩¹, 马晓娜¹, 高永艳², 刘毅¹, 马宏伟¹, 杨东强¹^,⁺()

1.山东建筑大学计算机科学与技术学院，济南 250101
2.山东建筑大学体育部，济南 250101

通讯作者: +E-mail: ydq@sdjzu.edu.cn
作者简介:王琳（1975—），女，山东泗水人，硕士研究生，讲师，主要研究方向为机器学习、智慧城市等。
孙倩（1983—），女，山东临沂人，硕士研究生，讲师，主要研究方向为智慧城市。
马晓娜（1977—），女，河北衡水人，硕士研究生，副教授，主要研究方向为软件测试、机器学习等。
高永艳（1980—），女，潍坊昌邑人，硕士研究生，副教授，主要研究方向为体育教学与训练、体育文化等。
刘毅（1978—），男,山东威海人，硕士研究生，讲师，主要研究方向为机器视觉、人工智能等
马宏伟(1969—)，男，山东济阳人，博士研究生，教授,硕士生导师，主要研究方向为计算机系统结构、计算机网络、机器学习等。
杨东强（1970—），男，山东招远人，博士研究生，副教授，硕士生导师，主要研究方向为自然语言处理、人工智能等。
基金资助:
山东省自然科学基金(ZR2018MF012);山东省自然科学基金(ZR2020QF029);山东建筑大学博士基金(XNBS1811)

Abstract

Abstract:

To solve the contradiction between multiple wearable sensor features and the limited computing power and storage capacity of embedded devices, feature engineering is used to select the best features for predicting physical activity energy expenditure (PAEE) on the basis of data fusion of multiple sensors (accelerometer and gyroscope sensors). In the data preprocessing stage, time-domain and frequency-domain features of the sensor are extracted by using sliding window technology, and sinusoidal curve fitting is used for dataset at three velocity levels, finally hypothesis testing is carried out to check data outliers. A WEKA experimental platform is constructed based on filtering, warpper and embedded feature selection algorithms and machine learning prediction models such as multiple linear regression, regression tree, support vector machine and neural network. Finally, the optimal model is selected by evaluating the correlation coefficient and mean absolute error of each model during the decision level fusion. The dataset with jitter is used as the test data, which shows that feature selection can mitigate model overfitting and improve the model’s generalization ability and robustness. Embedded feature selection adopts classical elastic network algorithm. Experimental results show that the features extracted from accelerometer sensors play a more decisive role than those from gyroscope sensors in PAEE and the neural network model of multi-sensor feature fusion based on correlation coefficient method is the optimal model.

Key words: physical activity energy expenditure, feature selection, neural network

摘要：

为了解决可穿戴传感器特征过多与嵌入式设备计算能力和存储能力有限的矛盾问题，在多传感器（加速度传感器、陀螺仪传感器）数据融合的基础上，采用特征工程的方法选出人体运动能量消耗预测（PAEE）的最优特征。在数据预处理阶段，使用滑动窗口技术提取传感器的时域、频域特征，对三个速度水平的数据集使用正弦曲线拟合，并通过显著性差异检验分析选出有效数据。构建了过滤式、封装式和嵌入式特征选择算法与多线性回归、回归树、支持向量机和神经网络等机器学习预测模型结合的WEKA实验平台。最后决策级融合时，通过评估每个模型的相关系数和平均绝对误差选择出最优模型。模型训练时采用带抖动的数据集作为测试集，避免出现模型的过拟合现象，提高模型的泛化能力和鲁棒性。嵌入式特征选择采用经典的弹性网络算法。实验结果表明，在PAEE中加速度计传感器的特征比陀螺仪传感器的特征更具有决定性的作用，基于相关系数方法的多传感器特征融合的神经网络模型是最优模型。

关键词: 人体活动能量消耗, 特征选择, 神经网络

CLC Number:

TP391

WANG Lin, SUN Qian, MA Xiaona, GAO Yongyan, LIU Yi, MA Hongwei, YANG Dongqiang. Research on Prediction Model of Physical Activity Energy Expenditure with Wearable Sensors[J]. Journal of Frontiers of Computer Science and Technology, 2022, 16(12): 2832-2840.

王琳, 孙倩, 马晓娜, 高永艳, 刘毅, 马宏伟, 杨东强. 可穿戴传感器的人体活动能量预测模型研究[J]. 计算机科学与探索, 2022, 16(12): 2832-2840.

Figures/Tables 15

Fig.1 Typical PAEE processing flow of sensor

Table 1 Experimental parameters

被试基本信息		传感器	速度水平/(km/h)	采样频率/Hz
人数	10人7男3女	臀部三轴加速度器HP_A	3.2	0.2
年龄/岁	49±12	脚踝三轴加速度器AK_A	4.8
身高/cm	178±10	臀部三轴陀螺仪HP_G	6.4
体重/kg	80.7±4.6	脚踝三轴陀螺仪AK_G

Table 2 Time-domains and frequency-domains characteristics of PAEE

特征分类	特征	计算方法
时域特征	Counts	$∑ i = 1 N x i$
	均值mean	$s - = 1 N ∑ i = 1 N s i$
	标准差SD	$σ s = 1 N ∑ i = 1 N s i - 1 N ∑ i = 1 N s i$
	变异系数CV	$σ s ∕ \| s - \|$
	最小值、最大值	Max、min
	分位点	第10、25、50、75、90分位点
	IQR	$Q 75 t h$ - $Q 25 t h$
频域特征	主频	通过快速傅里叶变换获得
	振幅	通过快速傅里叶变换获得
	光谱熵	$- ∑ p (f k) l g p (f k)$

Table 2 Time-domains and frequency-domains characteristics of PAEE

特征分类	特征	计算方法
时域特征	Counts	$∑ i = 1 N x i$
	均值mean	$s - = 1 N ∑ i = 1 N s i$
	标准差SD	$σ s = 1 N ∑ i = 1 N s i - 1 N ∑ i = 1 N s i$
	变异系数CV	$σ s ∕ \| s - \|$
	最小值、最大值	Max、min
	分位点	第10、25、50、75、90分位点
	IQR	$Q 75 t h$ - $Q 25 t h$
频域特征	主频	通过快速傅里叶变换获得
	振幅	通过快速傅里叶变换获得
	光谱熵	$- ∑ p (f k) l g p (f k)$

Table 3 Significant difference test results

被试者	组内数据结果（F值）
被试者	HAcount	HGcount	AAcount	AGcount
被试1	13.165 350	0.646 716	1.983 844	0.635 209
被试2	4.269 025	0.089 645	1.974 077	0.858 583

Table 4 Comparison of evaluation results before and after feature selection (correlation coefficient)

比较策略		评估结果
选择方法		CfsSubsetEval+bestFirst^[22]
选取特征		6，7，68，79，83，87∶6
回归方法		多线性回归
特征选择前	CC	0.291 1
特征选择前	MAE	1.142 0
特征选择后	CC	0.829 4
特征选择后	MAE	0.533 8

Fig.2 Neural network model connection diagram

Fig.3 Evaluation results of correlation coefficient selection feature

Table 5 Comparison of evaluation results before and after feature selection (MI)

比较策略		评估结果
选择方法		互信息
选取特征		3，21，27，7，6，2，69，26，114，12，28，70，68，125，61，120，10，112，22，123∶20
回归方法		多线性回归
特征选择前	CC	0.417 7
特征选择前	MAE	0.428 2
特征选择后	CC	0.905 5
特征选择后	MAE	0.399 9

Fig.4 Comparison chart of evaluation results of mutual information selection feature

Table 6 Comparison of evaluation results before and after feature selection (wrapper)

比较策略		评估结果
选择方法		包装（wrapper）
选取特征		3，21，27，7，6，2，69，26，114，12，28，70，68，125，61，120，10，112，22，123:20
支持向量机方法		RBF+ranker
特征选择前	CC	0.291 1
特征选择前	MAE	1.142 0
特征选择后	CC	0.860 8
特征选择后	MAE	0.470 4

Fig.5 Comparison chart of evaluation results of wrapper selection feature

Table 7 Evaluation results of embedded method (elastic network)

方法	参数	CC	MAE
ElasticNET	$α = 0.3785$	0.783 4	0.561 6

Table 7 Evaluation results of embedded method (elastic network)

方法	参数	CC	MAE
ElasticNET	$α = 0.3785$	0.783 4	0.561 6

Fig.6 Comparison chart of evaluation results of optimal combination method

Table 8 Summary of evaluation results of optimal combination methods

特征选择方法	机器学习算法	CC	MAE
相关系数方法	神经网络	0.870 1	0.453 4
互信息方法	支持向量机	0.739 4	0.542 8
包装方法	支持向量机	0.860 8	0.470 4
嵌入式方法	弹性网络	0.783 8	0.560 1

Fig.7 Proposed regression system for PAEE

References 28

[1]	LIU S P, GAO R X, FREEDSON P S. Computational methods for estimating energy expenditure in human physical activities[J]. Medicine and Science in Sports and Exercise, 2012, 44(11): 2138-2146. DOI PMID
[2]	MIGUELES J H, CADENAS-SANCHEZ C, EKELUND U, et al. Accelerometer data collection and processing criteria to assess physical activity and other outcomes: a systematic review and practical considerations[J]. Sports Medicine, 2017, 47(9): 1821-1845. DOI PMID
[3]	范江江, 陈庆果. 加速度计测量体力活动的算法研究进展[J]. 湖北体育科技, 2016, 35(7): 595-598.
	FAN J J, CHEN Q G. Study of the calculating methods on physical activities with acceleration[J]. Hubei Sports Science, 2016, 35(7): 595-598.
[4]	陈庆果, 袁川, 范江江, 等. 智能手机内置加速度传感器监测走跑运动能量消耗的研究[J]. 首都体育学院学报, 2018, 30(5): 473-480.
	CHEN Q G, YUAN C, FAN J J, et al. Assessment of energy expenditure of walking and running using built-in accelerometer of smartphones[J]. Journal of Capital University of Physical Education and Sports, 2018, 30(5): 473-480.
[5]	陈庆果, 李翔. 佩戴部位对加速度计能耗监测准确性的影响: 算法的调节效应[J]. 中国体育科技, 2019, 55(3): 73-81.
	CHEN Q G, LI X. Effects of wearing sites of accelerometer on accuracy prediction of energy expenditure: moderating effect of algorithm[J]. China Sport Science and Technology, 2019, 55(3): 73-81.
[6]	HAMID A. Predicting children’s energy expenditure during physical activity using deep learning and wearable sensor data[J]. European Journal of Sport Science, 2021, 21(6): 918-926. DOI URL
[7]	钱慧芳, 易剑平, 付云虎. 基于深度学习的人体动作识别综述[J]. 计算机科学与探索, 2021, 15(3): 438-455. DOI URL
	QIAN H F, YI J P, FU Y H. Review of human action recognition based on deep learning[J]. Journal of Frontiers of Computer Science and Technology, 2021, 15(3): 438-455. DOI URL
[8]	陆姣姣, 邱俊. 加速度计在能量消耗预测中的应用研究综述[J]. 南京体育学院学报, 2014, 13(4): 24-29.
	LU J J, QIU J. A review of accelerometer for activity energy expenditure estimating[J]. Journal of Nanjing Sports Institute, 2014, 13(4): 24-29.
[9]	SCHULDHAUS D, DORN S, LEUTHEUSER H, et al. An adaptable inertial sensor fusion-based approach for energy expenditure estimation[C]// Proceedings of the 15th International Conference on Biomedical Engineering, Singapore, Dec 4-7, 2013. Berlin, Heidelberg: Springer, 2013: 124-127.
[10]	赵蕊. 基于WEKA平台的决策树算法设计与实现[D]. 长沙: 中南大学, 2007.
	ZHAO R. Design and implementation of decision tree algorithm based on WEKA platform[D]. Changsha: Central South University, 2007.
[11]	袁梅宇. 数据挖掘与机器学习——WEKA应用技术与实践[M]. 2版. 北京: 清华大学出版社, 2019.
	YUAN M Y. Data mining and machine learning—WEKA application technology and practice[M]. 2nd ed. Beijing: Tsinghua University Press, 2019.
[12]	黄仁, 田丰, 田维兴. 基于加速度传感器的运动模式识别[J]. 计算机工程与应用, 2015, 51(6): 235-239.
	HUANG R, TIAN F, TIAN W X. Motion pattern recognition using acceleration transducer[J]. Computer Engineering and Applications, 2015, 51(6): 235-239.
[13]	张昕阳. 运动方式识别和能量消耗模型的研究[D]. 上海: 华东师范大学, 2015.
	ZHANG X Y. Research on motion pattern recognition and energy consumption model[D]. Shanghai: East China Normal University, 2015.
[14]	梁志国, 孟晓风. 正弦波形参数拟合方法述评[J]. 测试技术学报, 2010, 24(1): 1-8.
	LIANG Z G, MENG X F. Review of sine wave curve-fit methods[J]. Journal of Test and Measurement Technology, 2010, 24(1): 1-8.
[15]	齐国清, 吕健. 正弦曲线拟合若干问题探讨[J]. 计算机工程与设计, 2008, 29(14): 3677-3680.
	QI G Q, LV J. Investigation of sine wave fitting algorithms[J]. Computer Engineering and Design, 2008, 29(14): 3677-3680.
[16]	姚旭, 王晓丹, 张玉玺, 等. 特征选择方法综述[J]. 控制与决策, 2012, 27(2): 161-166.
	YAO X, WANG X D, ZHANG Y X, et al. Summary of feature selection algorithms[J]. Control and Decision, 2012, 27(2): 161-166.
[17]	HANCER E, XUE B, ZHANG M J. A survey on feature selection approaches for clustering[J]. Artificial Intelligence Review, 2020, 53(6): 4519-4545. DOI URL
[18]	刘毛溪, 万鸣华, 孙成立, 等. 无监督的稀疏差分嵌入特征提取方法[J]. 小型微型计算机系统, 2017, 38(5): 1134-1138.
	LIU M X, WAN M H, SUN C L, et al. Unsupervised sparse difference embedding for feature extraction[J]. Journal of Chinese Computer Systems, 2017, 38(5): 1134-1138.
[19]	田冰. 经典统计学与机器学习中变量选择方法的比较分析[D]. 济南: 山东大学, 2019.
	TIAN B. Comparative analysis of variable selection methods in classical statistics and machine learning[D]. Jinan: Shandong University, 2019.
[20]	GJORESKI H, KALUŽA B, GAMS M, et al. Context-based ensemble method for human energy expenditure estimation[J]. Applied Soft Computing, 2015, 37: 960-970. DOI URL
[21]	MONTOYE A H K, BEGUM M, HENNING Z, et al. Comparison of linear and non-linear models for predicting energy expenditure from raw accelerometer data[J]. Physiological Measurement, 2017, 38(2): 343-357. DOI PMID
[22]	郑增威, 杜俊杰, 霍梅梅, 等. 基于可穿戴传感器的人体活动识别研究综述[J]. 计算机应用, 2018, 38(5): 1223-1229. DOI
	ZHENG Z W, DU J J, HUO M M, et al. Review of human activity recognition based on wearable sensors[J]. Journal of Computer Applications, 2018, 38(5): 1223-1229. DOI
[23]	DE CRAEMER M, DE DECKER E, SANTOS-LOZANO A, et al. Validity of the Omron pedometer and the actigraph step count function in preschoolers[J]. Journal of Science and Medicine in Sport, 2015, 18(3): 289-293. DOI PMID
[24]	卢颖. 广义线性模型基于Elastic Net的变量选择方法研究[D]. 北京: 北京交通大学, 2011: 86-123.
	LU Y. Research on variable selection method for generalized linear models based on Elastic Net[D]. Beijing: Beijing Jiaotong University, 2011: 86-123.
[25]	樊振宇. BP神经网络模型与学习算法[J]. 软件导刊, 2011, 10(7): 66-68.
	FAN Z Y. BP neural network model and learning algorithm[J]. Software Guide, 2011, 10(7): 66-68.
[26]	FAN J, LV J. A selective overview of variable selection in high dimensional feature space[J]. Statistical Sinica, 2010, 20(1): 101-148.
[27]	王三保, 陈浩. 人体运动能量消耗测量的方法学分析[J]. 体育时空, 2013(9): 128-129.
	WANG S B, CHEN H. Methodological analysis of measurement of human sports energy consumption[J]. Sports Time and Space, 2013(9): 128-129.
[28]	ZHANG C H. Nearly unbiased variable selection under minimax concave penalty[J]. The Annals of Statistics, 2010, 38(2): 894-942.

Research on Prediction Model of Physical Activity Energy Expenditure with Wearable Sensors

可穿戴传感器的人体活动能量预测模型研究

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 15

References 28

Related Articles 15

Recommended Articles

Metrics

[1]	SUN Shuifa, LI Xiaolong, LI Weisheng, LEI Dajiang, LI Sihui, YANG Liu, WU Yirong. Review of Graph Neural Networks Applied to Knowledge Graph Reasoning [J]. Journal of Frontiers of Computer Science and Technology, 2023, 17(1): 27-52.
[2]	WANG Shichen, HUANG Kai, CHEN Zhigang, ZHANG Wendong. Survey on 3D Human Pose Estimation of Deep Learning [J]. Journal of Frontiers of Computer Science and Technology, 2023, 17(1): 74-87.
[3]	PI Hong, LUO Chuan, LI Tianrui, CHEN Hongmei. Improved Fuzzy Rank Mutual Information for Monotonic Feature Selection [J]. Journal of Frontiers of Computer Science and Technology, 2023, 17(1): 108-115.
[4]	MA Fuyuan, WANG Ying, LI Lina, WANG Hongji. Structure and Feature Fusion Graph Hierarchical Pooling Model [J]. Journal of Frontiers of Computer Science and Technology, 2023, 17(1): 179-186.
[5]	GAO Yang, LIU Yuan. Recommendation Algorithm Combining Social Relationship and Knowledge Graph [J]. Journal of Frontiers of Computer Science and Technology, 2023, 17(1): 238-250.
[6]	LYU Xiaoqi, JI Ke, CHEN Zhenxiang, SUN Runyuan, MA Kun, WU Jun, LI Yidong. Expert Recommendation Algorithm Combining Attention and Recurrent Neural Network [J]. Journal of Frontiers of Computer Science and Technology, 2022, 16(9): 2068-2077.
[7]	ZHU Bingyu, LIU Zhen, ZHANG Jingxiang. COVID-19 Detection Algorithm Combining Grad-CAM and Convolutional Neural Network [J]. Journal of Frontiers of Computer Science and Technology, 2022, 16(9): 2108-2120.
[8]	AN Fengping, LI Xiaowei, CAO Xiang. Medical Image Classification Algorithm Based on Weight Initialization-Sliding Window CNN [J]. Journal of Frontiers of Computer Science and Technology, 2022, 16(8): 1885-1897.
[9]	HUANG Hao, GE Hongwei. Deep Residual Expression Recognition Network to Enhance Inter-class Discrimination [J]. Journal of Frontiers of Computer Science and Technology, 2022, 16(8): 1842-1849.
[10]	YU Huilin, CHEN Wei, WANG Qi, GAO Jianwei, WAN Huaiyu. Knowledge Graph Link Prediction Based on Subgraph Reasoning [J]. Journal of Frontiers of Computer Science and Technology, 2022, 16(8): 1800-1808.
[11]	LI Yuxuan, HONG Xuehai, WANG Yang, TANG Zhengzheng, BAN Yan. Groupwise Learning to Rank Algorithm with Introduction of Activated Weighting [J]. Journal of Frontiers of Computer Science and Technology, 2022, 16(7): 1594-1602.
[12]	ZHANG Yancao, ZHAO Yuhai, SHI Lan. Multi-feature Based Link Prediction Algorithm Fusing Graph Attention [J]. Journal of Frontiers of Computer Science and Technology, 2022, 16(5): 1096-1106.
[13]	OU Yangliu, HE Xi, QU Shaojun. Fully Convolutional Neural Network with Attention Module for Semantic Segmentation [J]. Journal of Frontiers of Computer Science and Technology, 2022, 16(5): 1136-1145.
[14]	CHENG Weiyue, ZHANG Xueqin, LIN Kezheng, LI Ao. Deep Convolutional Neural Network Algorithm Fusing Global and Local Features [J]. Journal of Frontiers of Computer Science and Technology, 2022, 16(5): 1146-1154.
[15]	TONG Gan, HUANG Libo. Review of Winograd Fast Convolution Technique Research [J]. Journal of Frontiers of Computer Science and Technology, 2022, 16(5): 959-971.