Use this link to cite:
http://hdl.handle.net/2183/31354 Análisis de movimiento mediante textil inteligente
Loading...
Identifiers
Publication date
Authors
Delgado-Oleas, Gabriel
Avellar, Leticia
Stefano Filho, Carlos Alberto
Frizera Neto, Anselmo
Leal Junior, Arnaldo
Rocón, Eduardo
Advisors
Other responsabilities
Journal Title
Bibliographic citation
Delgado-Oleas, G., Avellar, L., Stefano-Filho, C., Frizera, A., Leal-Junior, A., Rocon, E. (2022) Análisis de movimiento mediante textil inteligente. XLIII Jornadas de Automática: libro de actas, pp.66-70 https://doi.org/10.17979/spudc.9788497498418.0066
Type of academic work
Academic degree
Abstract
[Resumen] Este artículo presenta el desarrollo de un textil inteligente (smart garment) portátil dotado de 30 sensores de fibra óptica multiplexados que, a través de algoritmos de Inteligencia Artificial (IA), es capaz de clasificar de actividades de múltiples sujetos. Se evalúan seis actividades diarias: de pie, sentado, en cuclillas, brazos arriba y abajo, caminar y correr. El clasificador utilizado es el de k vecinos más cercanos y los resultados de 10 ensayos de todos los voluntarios presentaron una precisión de 94,00 (0,14) %. La cadencia y la frecuencia respiratoria se estimaron y compararon con los datos de una unidad de medición inercial ubicada en la parte posterior de la prenda. El error más alto fue del 2,22 %. El enfoque propuesto presentó la viabilidad para el reconocimiento de actividad y la extracción de parámetros relacionados con el movimiento, en el ámbito de los desarrollos en Healthcare 4.0.
[Abstract] This paper presents the development of a wearable smart garment equipped with 30 multiplexed fiber optic sensors that, through Artificial Intelligence (AI) algorithms, can classify multiple subjects' activities. Six daily activities are evaluated: standing, sitting, squatting, arms up and down, walking, and running. The classifier used is k-nearest neighbors, and results from 10 trials of all volunteers showed an accuracy of 94.00 (0.14)%. Cadence and respiratory rate were estimated and compared with data from an inertial measurement unit located on the back of the garment. The highest error was 2.22%. The proposed approach presented feasibility for activity recognition and extraction of motion-related parameters in the scope of Healthcare 4.0 developments.
[Abstract] This paper presents the development of a wearable smart garment equipped with 30 multiplexed fiber optic sensors that, through Artificial Intelligence (AI) algorithms, can classify multiple subjects' activities. Six daily activities are evaluated: standing, sitting, squatting, arms up and down, walking, and running. The classifier used is k-nearest neighbors, and results from 10 trials of all volunteers showed an accuracy of 94.00 (0.14)%. Cadence and respiratory rate were estimated and compared with data from an inertial measurement unit located on the back of the garment. The highest error was 2.22%. The proposed approach presented feasibility for activity recognition and extraction of motion-related parameters in the scope of Healthcare 4.0 developments.
Description
Editor version
Rights
Atribución-NoComercial-CompartirIgual 4.0 Internacional (CC BY-NC-SA 4.0)
https://creativecommons.org/licenses/by-nc-sa/4.0/deed.es


