Predicting energy cost from wearable sensors: A dataset of energetic and physiological wearable sensor data from healthy individuals performing multiple physical activities

This dataset presents energetic and wearable physiological sensor data from ten healthy subjects performing six physical activities.<br>The activities tested were: walking, incline walking, backwards walking, and running on a treadmill, cycling on a stationary bike, and stair climbing on a stairmill -- all at a variety of speeds and/or intensities (21 total conditions). The following physiological signals were collected from wearable sensors while subjects performed all the activities: - Oxygen consumption and carbon dioxide production- Respiratory exchange ratio- Breath frequency- Minute ventilation - Oxygen saturation (SpO₂)- Heart rate- Electrodermal activity- Skin temperature - Accelerations, angular velocity, and magnetic field measured from left/right wrist, left/right ankle, left/right foot, pelvis, and chest (IMUs)- Surface EMG from left/right gluteus maximus, rectus femoris, vastus lateralis, semitendinosis, biceps femoris, medial gastrocnemius, soleus, tibialis anterior<br>The data are contained in ten (10) Matlab .mat files (one for each subject). For a complete description of the file structure please see the file: CompleteDataDescription_Ingraham_Ferris_Remy_2018<br>For a complete description of experimental methods please see the published article: Ingraham, Kimberly A., Daniel P. Ferris, and C. David Remy. "Evaluating Physiological Signal Salience for Estimating Metabolic Energy Cost from Wearable Sensors." <i>Journal of Applied Physiology</i> (2019). DOI: 10.1152/japplphysiol.00714.2018<br><br>Edit history: Version 4 is the most current version (as of 3/12/2019). The only changes made between versions were updates to the CompleteDataDescription.pdf file for completeness. <br><br>Please direct any correspondence to: Kimberly Ingraham ([e-mail])<br><br><br><br>

本数据集收录了10名健康受试者完成6种体力活动时的能量代谢相关及穿戴式生理传感器采集数据。
本次测试涵盖的活动包括：平地行走、上坡行走、倒退行走、跑步机跑步、固定式自行车骑行及爬梯机爬楼，所有活动均设置了多种速度与/或强度，共计21种实验条件。
受试者完成全部活动期间，穿戴式传感器采集了如下生理信号：
- 耗氧量与二氧化碳生成量
- 呼吸交换率
- 呼吸频率
- 每分通气量
- 血氧饱和度（SpO₂）
- 心率
- 皮肤电活动
- 皮肤温度
- 左右手腕、左右脚踝、左右足部、骨盆及胸部采集的加速度、角速度与磁场强度（惯性测量单元（IMU）数据）
- 左右侧臀大肌、股直肌、股外侧肌、半腱肌、股二头肌、腓肠肌内侧头、比目鱼肌及胫骨前肌的表面肌电信号
数据集以10个Matlab格式的.mat数据文件存储，每名受试者对应一个文件。如需完整了解文件结构，请参阅文件：《CompleteDataDescription_Ingraham_Ferris_Remy_2018》。
如需完整了解实验方法，请参阅已发表论文：Ingraham, Kimberly A., Daniel P. Ferris, 及 C. David Remy. 《基于穿戴式传感器估算代谢能耗的生理信号显著性评估》，*Journal of Applied Physiology*（2019）。DOI: 10.1152/japplphysiol.00714.2018
修订历史：截至2019年3月12日，当前最新版本为V4。各版本间仅对《CompleteDataDescription.pdf》文件进行了完善性更新。
如有任何通信或疑问，请联系：Kimberly Ingraham（邮箱：[e-mail]）

搜集汇总

背景与挑战

背景概述

该数据集包含10名健康受试者进行6种体力活动(如步行、跑步、骑行等)时的多模态生理传感器数据，涵盖氧气消耗、心率、肌电信号等多种指标，旨在研究可穿戴传感器预测能量消耗的能力。数据以Matlab格式存储，每个受试者单独一个文件。