Monitoring floor vibrations to evaluate objectively physical activity during housework activities

Tripette, Julien; Sasaki, M.; Kuno-Mizumura, Mayumi; Motooka, Nobuhisa; Ohta, Y.

doi:10.1109/lifetech52111.2021.9391942

Cited by 3 publications

(1 citation statement)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Step count-data from the uniformized vector norm time series were cut in windows of 1 second with an overlap of 50%. For each window, the number of steps was computed using a standard peak detection algorithm configured to detect vibration peaks with a minimum prominence of 1.05 SD and a minimum interval of 250 ms [27]. The step count parameter was computed for each activity of each participant by summing the number of unique peaks throughout the activity (6 min).…”

mentioning

confidence: 99%

Objective Assessment of Physical Activity at Home Using a Novel Floor-Vibration Monitoring System: Validation and Comparison With Wearable Activity Trackers and Indirect Calorimetry Measurements

Nakajima,

Kitayama,

Ohta

et al. 2024

JMIR Form Res

Self Cite

View full text Add to dashboard Cite

Background The self-monitoring of physical activity is an effective strategy for promoting active lifestyles. However, accurately assessing physical activity remains challenging in certain situations. This study evaluates a novel floor-vibration monitoring system to quantify housework-related physical activity. Objective This study aims to assess the validity of step-count and physical behavior intensity predictions of a novel floor-vibration monitoring system in comparison with the actual number of steps and indirect calorimetry measurements. The accuracy of the predictions is also compared with that of research-grade devices (ActiGraph GT9X). Methods The Ocha-House, located in Tokyo, serves as an independent experimental facility equipped with high-sensitivity accelerometers installed on the floor to monitor vibrations. Dedicated data processing software was developed to analyze floor-vibration signals and calculate 3 quantitative indices: floor-vibration quantity, step count, and moving distance. In total, 10 participants performed 4 different housework-related activities, wearing ActiGraph GT9X monitors on both the waist and wrist for 6 minutes each. Concurrently, floor-vibration data were collected, and the energy expenditure was measured using the Douglas bag method to determine the actual intensity of activities. Results Significant correlations (P<.001) were found between the quantity of floor vibrations, the estimated step count, the estimated moving distance, and the actual activity intensities. The step-count parameter extracted from the floor-vibration signal emerged as the most robust predictor (r2=0.82; P<.001). Multiple regression models incorporating several floor-vibration–extracted parameters showed a strong association with actual activity intensities (r2=0.88; P<.001). Both the step-count and intensity predictions made by the floor-vibration monitoring system exhibited greater accuracy than those of the ActiGraph monitor. Conclusions Floor-vibration monitoring systems seem able to produce valid quantitative assessments of physical activity for selected housework-related activities. In the future, connected smart home systems that integrate this type of technology could be used to perform continuous and accurate evaluations of physical behaviors throughout the day.

show abstract

mentioning

confidence: 99%

Objective Assessment of Physical Activity at Home Using a Novel Floor-Vibration Monitoring System: Validation and Comparison With Wearable Activity Trackers and Indirect Calorimetry Measurements

Nakajima,

Kitayama,

Ohta

et al. 2024

JMIR Form Res

Self Cite

View full text Add to dashboard Cite

show abstract

A Machine Learning Approach for Predicting Physical Activity Intensity from Wearable Sensor Data

Lavanya,

2024

2024 7th International Conference on Circuit Power and Computing Technologies (ICCPCT)

View full text Add to dashboard Cite

Objective Assessment of Physical Activity at Home Using a Novel Floor-Vibration Monitoring System: Validation and Comparison With Wearable Activity Trackers and Indirect Calorimetry Measurements (Preprint)

Nakajima,

Kitayama,

Ohta

et al. 2023

Preprint

Self Cite

View full text Add to dashboard Cite

BACKGROUND The self-monitoring of physical activity is an effective strategy for promoting active lifestyles. However, accurately assessing physical activity remains challenging in certain situations. This study evaluates a novel floor-vibration monitoring system to quantify housework-related physical activity. OBJECTIVE This study aims to assess the validity of step-count and physical behavior intensity predictions of a novel floor-vibration monitoring system in comparison with the actual number of steps and indirect calorimetry measurements. The accuracy of the predictions is also compared with that of research-grade devices (ActiGraph GT9X). METHODS The Ocha-House, located in Tokyo, serves as an independent experimental facility equipped with high-sensitivity accelerometers installed on the floor to monitor vibrations. Dedicated data processing software was developed to analyze floor-vibration signals and calculate 3 quantitative indices: floor-vibration quantity, step count, and moving distance. In total, 10 participants performed 4 different housework-related activities, wearing ActiGraph GT9X monitors on both the waist and wrist for 6 minutes each. Concurrently, floor-vibration data were collected, and the energy expenditure was measured using the Douglas bag method to determine the actual intensity of activities. RESULTS Significant correlations (P<.001) were found between the quantity of floor vibrations, the estimated step count, the estimated moving distance, and the actual activity intensities. The step-count parameter extracted from the floor-vibration signal emerged as the most robust predictor (r2=0.82; P<.001). Multiple regression models incorporating several floor-vibration–extracted parameters showed a strong association with actual activity intensities (r2=0.88; P<.001). Both the step-count and intensity predictions made by the floor-vibration monitoring system exhibited greater accuracy than those of the ActiGraph monitor. CONCLUSIONS Floor-vibration monitoring systems seem able to produce valid quantitative assessments of physical activity for selected housework-related activities. In the future, connected smart home systems that integrate this type of technology could be used to perform continuous and accurate evaluations of physical behaviors throughout the day.

show abstract

Monitoring floor vibrations to evaluate objectively physical activity during housework activities

Cited by 3 publications

References 14 publications

Objective Assessment of Physical Activity at Home Using a Novel Floor-Vibration Monitoring System: Validation and Comparison With Wearable Activity Trackers and Indirect Calorimetry Measurements

Objective Assessment of Physical Activity at Home Using a Novel Floor-Vibration Monitoring System: Validation and Comparison With Wearable Activity Trackers and Indirect Calorimetry Measurements

A Machine Learning Approach for Predicting Physical Activity Intensity from Wearable Sensor Data

Objective Assessment of Physical Activity at Home Using a Novel Floor-Vibration Monitoring System: Validation and Comparison With Wearable Activity Trackers and Indirect Calorimetry Measurements (Preprint)

Contact Info

Product

Resources

About