PHYTMO database (doi:10.21950/7YRNMV)

View:

Part 1: Document Description
Part 2: Study Description
Part 5: Other Study-Related Materials
Entire Codebook

PHYTMO database

doi:10.21950/7YRNMV

e-cienciaDatos

2024-01-30

2

García-de-Villa, Sara; Jiménez Martín, Ana; García Domínguez, Juan Jesús, 2024, "PHYTMO database", https://doi.org/10.21950/7YRNMV, e-cienciaDatos, V2

PHYTMO database

doi:10.21950/7YRNMV

García-de-Villa, Sara (Universidad Rey Juan Carlos)

Jiménez Martín, Ana (Universidad de Alcalá. Departamento de Electrónica)

García Domínguez, Juan Jesús (Universidad de Alcalá. Departamento de Electrónica)

MATLAB R2020b for the syncronization of the IMU and optical data. To fill the gaps caused by occlusions in the data recording

FrailCheck SBPLY/17/180501/000392

MICROCEBUS RTI2018-095168-B-C51

e-cienciaDatos

Jiménez Martín, Ana

Jiménez Martín, Ana

2024-01-26

https://doi.org/10.21950/7YRNMV

Otra, Motion recognition, Performance evaluation, Inertial measurement units, IMU, Machine learning, ML, Elderly, Virtual coach

<p>Database of physical therapy exercises with variability of execution collected by wearable sensors. The PHYTMO database contains data from physical therapy exercises and gait variations recorded with magneto-inertial sensors, including information from an optical reference system. It includes the recording of 30 volunteers, aged between 20 and 70 years old. A total amount of 6 exercises and 3 gait variations commonly prescribed in physical therapies were recorded.</p> <p>Home-based physical therapies are specially important for older adults, however they are effective if the prescribed exercises are correctly executed and patients adhere to these routines. Older adults can easily forget the guidelines from therapists, so in this work, we propose the use of Machine Learning techniques to recognize which exercise is being carried out and to assess if the recognized exercise is properly executed by using data from four IMUs placed on the person limbs.</p>

2020-11-01-2021-05-01

Sara García-de-Villa, Ana Jiménez-Martín, & Juan Jesús García-Domínguez. (2022). A database of physical therapy exercises with variability of execution collected by wearable sensors (V0.1) [Data set]. Zenodo. https://doi.org/10.5281/zenodo.6319979

García-de-Villa, S., Jiménez-Martín, A. & García-Domínguez, J.J. A database of physical therapy exercises with variability of execution collected by wearable sensors. Sci Data 9, 266 (2022). https://doi.org/10.1038/s41597-022-01387-2

García-de-Villa, S., Jiménez-Martín, A. & García-Domínguez, J.J. A database of physical therapy exercises with variability of execution collected by wearable sensors. Sci Data 9, 266 (2022). https://doi.org/10.1038/s41597-022-01387-2

Sara García-de-Villa, David Casillas-Pérez, Ana Jiménez-Martín, Juan Jesús García-Domínguez. Simultaneous exercise recognition and evaluation in prescribed routines: Approach to virtual coaches. Expert Systems with Applications 199, 116990, (2022). https://doi.org/10.1016/j.eswa.2022.116990

Sara García-de-Villa, David Casillas-Pérez, Ana Jiménez-Martín, Juan Jesús García-Domínguez. Simultaneous exercise recognition and evaluation in prescribed routines: Approach to virtual coaches. Expert Systems with Applications 199, 116990, (2022). https://doi.org/10.1016/j.eswa.2022.116990

Label:

PHYTMO.zip

Text:

A total of 7 076 CSV files are included. Files are called with the nomenclature GNNEEELP_S, where G refers to the letter of the range of age, so it is “A”, “B”, “C”, “D” or “E”; NN is number of identification of the volunteer, which ranges from “01” to “10”; EEE indicates the type of exercise (KFE, HAA, SQT, EAH, EFE or SQZ) or gait variation (GAT, GIS or GHT); L is the leg that performed the exercise, so this letter is only included in the KFE and HAA exercises and it can be “L” or “R”; P is a label that indicates the evaluation of the exercise performance, which takes the “0” value when the file contains the correctly performed exercise and “1” when exercises are wrongly performed; and finally, S indicates the index of the series, being “1” for the first recorded series and “2” for the second one. The main directory includes three folders called “inertial”, “optical raw” and “optical interp”, which contain the data recorded with the IMUs and with the optical system, respectively. The “raw” and the “interp” folders contain the raw and interpolated data, respectively. The “inertial” folder is divided into two directories which refer to the two possible group of limbs, that is “upper” or “lower”. The corresponding internal structure is schematized in Fig. 5 and detailed in the following. Each limb directory contains five folders, corresponding to each age group of volunteers (“A” 20-29 y.o., “B” 30-39 y.o., “C” 40-49 y.o., “D” 50-59 y.o. and “E” older than 60 y.o.). The "optical raw” and “optical interp” folders include two directories: “biomech_model” and “rigid_bodies”. The data of the markers placed directly on the body of each subject are contained in the “biomech_model” folder and the data of the IMU mounting boards are located in the “rigid_bodies” folder.

application/zip

Label:

Readme-en_Phytmo_Garcia.txt

text/plain