Classification of daily activities using a wireless instrumented insole (WalkinSense) in a semi free-living setting
4 affiliations
- 1Department of Precision Health, Luxembourg Institute of Health, Strassen, Luxembourg
- 2Luxembourg Institute of Research in Orthopedics, Sports Medicine and Science, Eich, Luxembourg
- 3IEE S.A., Bissen, Luxembourg
- 4Centre Hospitalier de Luxembourg, Eich, Luxembourg
Abstract
Background: Accurate monitoring of activities of daily living (ADLs) in real‑world environments is essential for preventive care and rehabilitation, yet it remains difficult to achieve outside controlled laboratory settings. Instrumented insoles provide a promising, unobtrusive solution for continuous monitoring. However, the use of multimodal systems integrating plantar pressure and inertial data in naturalistic conditions and larger cohorts is still limited. This study therefore aims to evaluate the accuracy of a wireless instrumented insole (WalkinSense) that fuses pressure and inertial sensor data to classify ADLs within a semi free-living setting. Methods: A total of 99 participants performed a broad set of indoor and outdoor activities. Frame-by-frame performance was compared to ground truth (direct observation) using overall accuracy, Cohen’s Kappa, precision, recall, F1-scores and a normalised confusion matrix. Agreement on total activity duration was assessed using mean absolute percentage error (MAPE) scores and Bland-Altman plots. Results: Activity classification showed almost perfect agreement (mean overall accuracy 0.87, mean Cohen’s Kappa 0.84). Excellent performance (F1 > 0.90) was achieved for sitting, walking with crutches and cycling, while standing, level and non-level walking showed good performance (F1 > 0.80). Most misclassifications occurred between level walking, hill walking and stairs. Duration-based analysis confirmed high accuracy for sitting, walking with crutches and cycling (MAPE ≤ 10%). Bland-Altman plots indicated overestimation of level walking and underestimation of hill and stair walking. Step count was highly accurate (MAPE < 5%), whereas stair count showed only reasonable accuracy (MAPE ≈ 26%). Conclusion: These findings demonstrate the system’s strong potential for real-world monitoring and classification of ADLs while also highlighting the need for improved detection of non-level walking activities.
Related
- Wireless, spatially controlled in-situ methotrexate drug encapsulation and release from PEDOT films
- Portable and wireless microfluidic impedance sensor functionalized with MnCoPBA@MoS2 nanocomposite screen-printed gold electrodes for on-site bacteria detection
- Stretchable Wireless Sensing-Computing Biointerfaces Enabled by High-Voltage Megahertz Electrochemical Transistors
- Design and implementation of a Wireless Power Transfer with double-sided LCC compensation for a farming robot
- A self-powered finger-worn wireless tracking interface for human-machine interaction
BibTeX
@article{backes2026classification,
title = {{Classification of daily activities using a wireless instrumented insole (WalkinSense) in a semi free-living setting}},
author = {Anne Backes and Melanie Eckelt and Jennifer Fayad and Valeria Serchi and Thomas Solignac and Tobias Meyer and Bernd Grimm and Caroline Mouton and Romain Seil and Laurent Malisoux},
journal = {PLOS One},
year = {2026},
month = sep,
doi = {10.1371/journal.pone.0357961},
url = {https://doi.org/10.1371/journal.pone.0357961},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Classification of daily activities using a wireless instrumented insole (WalkinSense) in a semi free-living setting
AU - Anne Backes
AU - Melanie Eckelt
AU - Jennifer Fayad
AU - Valeria Serchi
AU - Thomas Solignac
AU - Tobias Meyer
AU - Bernd Grimm
AU - Caroline Mouton
AU - Romain Seil
AU - Laurent Malisoux
PY - 2026
DA - 2026/09/16
JO - PLOS One
DO - 10.1371/journal.pone.0357961
UR - https://doi.org/10.1371/journal.pone.0357961
AB - Background: Accurate monitoring of activities of daily living (ADLs) in real‑world environments is essential for preventive care and rehabilitation, yet it remains difficult to achieve outside controlled laboratory settings. Instrumented insoles provide a promising, unobtrusive solution for continuous monitoring. However, the use of multimodal systems integrating plantar pressure and inertial data in naturalistic conditions and larger cohorts is still limited. This study therefore aims to evaluate the accuracy of a wireless instrumented insole (WalkinSense) that fuses pressure and inertial sensor data to classify ADLs within a semi free-living setting. Methods: A total of 99 participants performed a broad set of indoor and outdoor activities. Frame-by-frame performance was compared to ground truth (direct observation) using overall accuracy, Cohen’s Kappa, precision, recall, F1-scores and a normalised confusion matrix. Agreement on total activity duration was assessed using mean absolute percentage error (MAPE) scores and Bland-Altman plots. Results: Activity classification showed almost perfect agreement (mean overall accuracy 0.87, mean Cohen’s Kappa 0.84). Excellent performance (F1 > 0.90) was achieved for sitting, walking with crutches and cycling, while standing, level and non-level walking showed good performance (F1 > 0.80). Most misclassifications occurred between level walking, hill walking and stairs. Duration-based analysis confirmed high accuracy for sitting, walking with crutches and cycling (MAPE ≤ 10%). Bland-Altman plots indicated overestimation of level walking and underestimation of hill and stair walking. Step count was highly accurate (MAPE < 5%), whereas stair count showed only reasonable accuracy (MAPE ≈ 26%). Conclusion: These findings demonstrate the system’s strong potential for real-world monitoring and classification of ADLs while also highlighting the need for improved detection of non-level walking activities.
SN - 1932-6203
ER -