Imperial College London
Portrait Picture

ProfessorDarioFarina

Faculty of Engineering, Department of Bioengineering

Chair in Neurorehabilitation Engineering
 
 
 
//

Contact

 

+44 (0)20 7594 1387d.farina Website

 
 
//

Location

 

RSM 4.15Royal School of MinesSouth Kensington Campus

//

Summary

 

Publications

Citation

BibTex format

@article{Pascual-Valdunciel:2023:10.3390/e25010114,
author = {Pascual-Valdunciel, A and Lopo-Martínez, V and Beltrán-Carrero, AJ and Sendra-Arranz, R and González-Sánchez, M and Pérez-Sánchez, JR and Grandas, F and Farina, D and Pons, JL and Oliveira, Barroso F and Gutiérrez, Á},
doi = {10.3390/e25010114},
journal = {Entropy (Basel, Switzerland)},
pages = {1--13},
title = {Classification of kinematic and electromyographic signals associated with pathological tremor using machine and deep learning.},
url = {http://dx.doi.org/10.3390/e25010114},
volume = {25},
year = {2023}
}
Download

RIS format (EndNote, RefMan)

TY  - JOUR
AB  - Peripheral Electrical Stimulation (PES) of afferent pathways has received increased interest as a solution to reduce pathological tremors with minimal side effects. Closed-loop PES systems might present some advantages in reducing tremors, but further developments are required in order to reliably detect pathological tremors to accurately enable the stimulation only if a tremor is present. This study explores different machine learning (K-Nearest Neighbors, Random Forest and Support Vector Machines) and deep learning (Long Short-Term Memory neural networks) models in order to provide a binary (Tremor; No Tremor) classification of kinematic (angle displacement) and electromyography (EMG) signals recorded from patients diagnosed with essential tremors and healthy subjects. Three types of signal sequences without any feature extraction were used as inputs for the classifiers: kinematics (wrist flexion-extension angle), raw EMG and EMG envelopes from wrist flexor and extensor muscles. All the models showed high classification scores (Tremor vs. No Tremor) for the different input data modalities, ranging from 0.8 to 0.99 for the f1 score. The LSTM models achieved 0.98 f1 scores for the classification of raw EMG signals, showing high potential to detect tremors without any processed features or preliminary information. These models may be explored in real-time closed-loop PES strategies to detect tremors and enable stimulation with minimal signal processing steps.
AU  - Pascual-Valdunciel,A
AU  - Lopo-Martínez,V
AU  - Beltrán-Carrero,AJ
AU  - Sendra-Arranz,R
AU  - González-Sánchez,M
AU  - Pérez-Sánchez,JR
AU  - Grandas,F
AU  - Farina,D
AU  - Pons,JL
AU  - Oliveira,Barroso F
AU  - Gutiérrez,Á
DO  - 10.3390/e25010114
EP  - 13
PY  - 2023///
SN  - 1099-4300
SP  - 1
TI  - Classification of kinematic and electromyographic signals associated with pathological tremor using machine and deep learning.
T2  - Entropy (Basel, Switzerland)
UR  - http://dx.doi.org/10.3390/e25010114
UR  - https://www.ncbi.nlm.nih.gov/pubmed/36673255
UR  - https://www.mdpi.com/1099-4300/25/1/114
UR  - http://hdl.handle.net/10044/1/101702
VL  - 25
ER  -
Download