Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: May 12, 2026

Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment
06:49

Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment

Published on: December 11, 2015

Activity recognition using a single accelerometer placed at the wrist or ankle.

Andrea Mannini1, Stephen S Intille, Mary Rosenberger

  • 11The BioRobotics Institute, Scuola Superiore Sant'Anna, Pisa, ITALY; 2College of Computer and Information Science and Bouvé College of Health Sciences, Northeastern University, Boston, MA; and 3Stanford Prevention Research Center, Stanford University, Stanford, CA.

Medicine and Science in Sports and Exercise
|April 23, 2013
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Diagnostic yield of electroencephalographyin the emergency department: protocol for the EMINENCE-M multicentre retrospective observational study.

BMJ open·2026
Same author

Perturbational complexity index detects subclinical cortical changes in early multiple sclerosis.

Clinical neurophysiology : official journal of the International Federation of Clinical Neurophysiology·2026
Same author

Do predictors of motor recovery differ between robotic and conventional post-stroke rehabilitation?

Journal of neuroengineering and rehabilitation·2026
Same author

Interpretable Machine Learning to Anticipate the Diagnostic Yield of EEG in the Emergency department. The EMINENCE study.

Journal of medical systems·2026
Same author

Tracheostomy weaning in patients with severe acquired brain injury: External validation of machine learning models.

Computer methods and programs in biomedicine·2026
Same author

Circulating Microglia-derived Extracellular Vesicles Predict Recovery After Rehabilitation in Stroke Survivors.

Translational stroke research·2026
Same journal

Are All Exertional Heat Strokes Equal?

Medicine and science in sports and exercise·2026
Same journal

Associations of 24-Hour Accelerometer-Measured Movement Behaviors with Chronic Liver Disease and the Mediating Role of Proteomics and Metabolomics.

Medicine and science in sports and exercise·2026
Same journal

Ultrasound Muscle Thickness Is a Poor Index of Criterion Magnetic Resonance Imaging Measures of Resistance Training-Induced Muscle Growth.

Medicine and science in sports and exercise·2026
Same journal

Heart Rate Variability During Exercise-Heat Stress Following Seven Days of Passive Heat Acclimation in Older Males: A Secondary Analysis.

Medicine and science in sports and exercise·2026
Same journal

Capillary Blood as a Potential New Matrix for the Athlete Biological Passport: A Field-Based Comparison Study.

Medicine and science in sports and exercise·2026
Same journal

Does Mental Fatigue Negatively Impact Physical Performance Fatiguability?

Medicine and science in sports and exercise·2026
See all related articles

This study developed an algorithm to classify physical activity from wrist and ankle accelerometer data. The algorithm accurately identifies activities like walking and cycling, with potential for real-time use on mobile devices.

Area of Science:

  • Wearable sensor technology
  • Biomedical engineering
  • Physical activity monitoring

Background:

  • Large-scale physical activity surveillance projects utilize wrist-worn accelerometers for raw data collection.
  • Increasing wear time by using wrist monitors instead of hip-worn devices is a key objective.
  • Improving activity type and intensity estimation from raw accelerometer signals is crucial.

Purpose of the Study:

  • To develop an algorithm for processing wrist and ankle accelerometer data.
  • To classify participant behavior into four broad activity classes: ambulation, cycling, sedentary, and other activities.

Main Methods:

  • Thirty-three participants wore wrist and ankle accelerometers during 26 daily activities.
  • Accelerometer data were preprocessed to extract features from 2-, 4-, and 12.8-second windows.

More Related Videos

A Method for Quantifying Upper Limb Performance in Daily Life Using Accelerometers
07:24

A Method for Quantifying Upper Limb Performance in Daily Life Using Accelerometers

Published on: April 21, 2017

Home-Based Monitor for Gait and Activity Analysis
07:24

Home-Based Monitor for Gait and Activity Analysis

Published on: August 8, 2019

Related Experiment Videos

Last Updated: May 12, 2026

Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment
06:49

Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment

Published on: December 11, 2015

A Method for Quantifying Upper Limb Performance in Daily Life Using Accelerometers
07:24

A Method for Quantifying Upper Limb Performance in Daily Life Using Accelerometers

Published on: April 21, 2017

Home-Based Monitor for Gait and Activity Analysis
07:24

Home-Based Monitor for Gait and Activity Analysis

Published on: August 8, 2019

  • A support vector machine classifier was trained using extracted features to identify activities, validated with a leave-one-subject-out strategy.

    • Main Results:

    • The algorithm achieved 95.0% accuracy for ankle data and 84.7% for wrist data using 12.8-second windows.
    • Using shorter 4-second windows minimally reduced wrist data accuracy to 84.2%.
    • The algorithm demonstrated high classification accuracy for four broad activity classes.

    Conclusions:

    • A 13-feature classification algorithm effectively categorizes physical activities from accelerometer data.
    • The algorithm is computationally efficient, enabling real-time implementation on mobile devices.
    • A latency of only 4 seconds is achievable for real-time activity classification.