Next Article in Journal
Towards Fingerprint Mosaicking Artifact Detection: A Self-Supervised Deep Learning Approach
Previous Article in Journal
One-Step Electrochemical Fabrication of Poly O-Cresolphthalein Complexone and Electrochemically Reduced Graphene Oxide Modified Electrode for Detection of Nitrofurantoin
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Sensors
Volume 26
Issue 12
10.3390/s26123683
sensors-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
This is an early access version, the complete PDF, HTML, and XML versions will be available soon.
Article

Impact of Gyroscope Integration, Sensor Placement, and Activity Granularity on Human Activity Recognition Performance

by
Alejandro Castellanos
1,*,
Antonio M. López
1,
Miguel Á. Salinas
1,
Juan C. Álvarez
1,
Diego Álvarez
1,
Gonzalo García
1,
Ángel Buendía-Romero
2,3,4,
Asier Mañas
2,3,5,
Raquel Bailón
6,
Vicente Martín
7,
Ana Carbonell-Baeza
8,
Verónica Cabanas-Sánchez
9,10 and
David Martinez-Gomez
9,10,11
1
Multisensor Systems and Robotics Research Group (SiMuR), Electrical Engineering Department, University of Oviedo, 33204 Gijón, Spain
2
GENUD Toledo Research Group, Faculty of Sports Sciences, Universidad de Castilla-La Mancha, 45071 Toledo, Spain
3
CIBER on Frailty and Healthy Aging (CIBERFES), Instituto de Salud Carlos III, 28029 Madrid, Spain
4
Grupo Mixto de Fragilidad y Envejecimiento Exitoso UCLM-SESCAM, Universidad de Castilla-La Mancha-Servicio de Salud de Castilla-La Mancha, IDISCAM, 45004 Toledo, Spain
5
Faculty of Education, Psychology and Sport Sciences, University of Huelva, 21007 Huelva, Spain
6
Biomedical Signal Interpretation and Computational Simulation Research Group (BSICoS), University of Zaragoza, Campus Río Ebro, I+D+i Building—C/Mariano Esquillor, S/N, 50018 Zaragoza, Spain
7
Gene-Environment-Health Interactions Research Group (GIIGAS), Department of Biomedical Sciences, Faculty of Veterinary Science, University of León, 24071 León, Spain
8
CTS-1038 eMpOwering health by physical actiVity, Exercise and nutrITion Research Group (MOVE-IT), Department of Physical Education, Instituto de Investigación e Innovación Biomédica de Cádiz (INiBICA), Universidad de Cádiz, 11002 Cadiz, Spain
9
Department of Preventive Medicine and Public Health, School of Medicine, Universidad Autonoma de Madrid, 28029 Madrid, Spain
10
CIBER of Epidemiology and Public Health, 28029 Madrid, Spain
11
IMDEA Nutrition, CEI UAM+CSIC, 28049 Madrid, Spain
 
add Show full affiliation list
 
remove Hide full affiliation list
*
Author to whom correspondence should be addressed.
Sensors 2026, 26(12), 3683; https://doi.org/10.3390/s26123683 (registering DOI)
Submission received: 10 March 2026 / Revised: 9 May 2026 / Accepted: 3 June 2026 / Published: 9 June 2026
(This article belongs to the Section Wearables)
Review Reports Versions Notes

Abstract

This study systematically evaluates the impact of sensor configuration, body location, classification granularity, and model choice on inertial-based human activity recognition in a laboratory dataset aligned with the Spanish IMPaCT cohort design. Data were collected from 85 participants instrumented with thigh-, wrist-, and hip-mounted inertial measurement units over a structured protocol of 13 semi-structured daily activities, a resting phase and a structured activity. After manual correction of timestamp drift, signals were segmented into overlapping 10-s windows and analyzed using convolutional neural networks, Random Forest, and XGBoost classifiers.Two classification targets were defined: fine-grained recognition of 15 laboratory-controlled activities and coarse-grained classification into four MET-based intensity levels. Results showed that classification granularity is the primary determinant of performance (F=224.85, p-value = 2.304×10−13 through the analysis of variance of the F1-score), with intensity-level classification substantially outperforming fine-grained activity recognition. Sensor configuration, model type, and body location also significantly influenced classification outcomes. Wrist-mounted sensors achieved the highest overall F1-scores. Incorporating gyroscope-derived features consistently improved performance across configurations, and feature importance analysis confirmed their substantial contribution. These findings, derived from models developed under controlled laboratory conditions, provide practical guidance for the design of wearable sensing protocols and modeling strategies in large-scale population-based studies, and support their extension to everyday physical activity, laying the foundation for future real-world applications.
Keywords: human activity recognition; inertial measurement unit; convolutional neural networks; decision tree-based models; physical activity; machine learning human activity recognition; inertial measurement unit; convolutional neural networks; decision tree-based models; physical activity; machine learning

Share and Cite

MDPI and ACS Style

Castellanos, A.; López, A.M.; Salinas, M.Á.; Álvarez, J.C.; Álvarez, D.; García, G.; Buendía-Romero, Á.; Mañas, A.; Bailón, R.; Martín, V.; et al. Impact of Gyroscope Integration, Sensor Placement, and Activity Granularity on Human Activity Recognition Performance. Sensors 2026, 26, 3683. https://doi.org/10.3390/s26123683

AMA Style

Castellanos A, López AM, Salinas MÁ, Álvarez JC, Álvarez D, García G, Buendía-Romero Á, Mañas A, Bailón R, Martín V, et al. Impact of Gyroscope Integration, Sensor Placement, and Activity Granularity on Human Activity Recognition Performance. Sensors. 2026; 26(12):3683. https://doi.org/10.3390/s26123683

Chicago/Turabian Style

Castellanos, Alejandro, Antonio M. López, Miguel Á. Salinas, Juan C. Álvarez, Diego Álvarez, Gonzalo García, Ángel Buendía-Romero, Asier Mañas, Raquel Bailón, Vicente Martín, and et al. 2026. "Impact of Gyroscope Integration, Sensor Placement, and Activity Granularity on Human Activity Recognition Performance" Sensors 26, no. 12: 3683. https://doi.org/10.3390/s26123683

APA Style

Castellanos, A., López, A. M., Salinas, M. Á., Álvarez, J. C., Álvarez, D., García, G., Buendía-Romero, Á., Mañas, A., Bailón, R., Martín, V., Carbonell-Baeza, A., Cabanas-Sánchez, V., & Martinez-Gomez, D. (2026). Impact of Gyroscope Integration, Sensor Placement, and Activity Granularity on Human Activity Recognition Performance. Sensors, 26(12), 3683. https://doi.org/10.3390/s26123683

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop