Cargando…

Monte Carlo Dropout for Uncertainty Estimation and Motor Imagery Classification

Motor Imagery (MI)-based Brain–Computer Interfaces (BCIs) have been widely used as an alternative communication channel to patients with severe motor disabilities, achieving high classification accuracy through machine learning techniques. Recently, deep learning techniques have spotlighted the stat...

Descripción completa

Detalles Bibliográficos
Autores principales: Milanés-Hermosilla, Daily, Trujillo Codorniú, Rafael, López-Baracaldo, René, Sagaró-Zamora, Roberto, Delisle-Rodriguez, Denis, Villarejo-Mayor, John Jairo, Núñez-Álvarez, José Ricardo
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2021
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8588128/
https://www.ncbi.nlm.nih.gov/pubmed/34770553
http://dx.doi.org/10.3390/s21217241
_version_ 1784598363820261376
author Milanés-Hermosilla, Daily
Trujillo Codorniú, Rafael
López-Baracaldo, René
Sagaró-Zamora, Roberto
Delisle-Rodriguez, Denis
Villarejo-Mayor, John Jairo
Núñez-Álvarez, José Ricardo
author_facet Milanés-Hermosilla, Daily
Trujillo Codorniú, Rafael
López-Baracaldo, René
Sagaró-Zamora, Roberto
Delisle-Rodriguez, Denis
Villarejo-Mayor, John Jairo
Núñez-Álvarez, José Ricardo
author_sort Milanés-Hermosilla, Daily
collection PubMed
description Motor Imagery (MI)-based Brain–Computer Interfaces (BCIs) have been widely used as an alternative communication channel to patients with severe motor disabilities, achieving high classification accuracy through machine learning techniques. Recently, deep learning techniques have spotlighted the state-of-the-art of MI-based BCIs. These techniques still lack strategies to quantify predictive uncertainty and may produce overconfident predictions. In this work, methods to enhance the performance of existing MI-based BCIs are proposed in order to obtain a more reliable system for real application scenarios. First, the Monte Carlo dropout (MCD) method is proposed on MI deep neural models to improve classification and provide uncertainty estimation. This approach was implemented using Shallow Convolutional Neural Network (SCNN-MCD) and with an ensemble model (E-SCNN-MCD). As another contribution, to discriminate MI task predictions of high uncertainty, a threshold approach is introduced and tested for both SCNN-MCD and E-SCNN-MCD approaches. The BCI Competition IV Databases 2a and 2b were used to evaluate the proposed methods for both subject-specific and non-subject-specific strategies, obtaining encouraging results for MI recognition.
format Online
Article
Text
id pubmed-8588128
institution National Center for Biotechnology Information
language English
publishDate 2021
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-85881282021-11-13 Monte Carlo Dropout for Uncertainty Estimation and Motor Imagery Classification Milanés-Hermosilla, Daily Trujillo Codorniú, Rafael López-Baracaldo, René Sagaró-Zamora, Roberto Delisle-Rodriguez, Denis Villarejo-Mayor, John Jairo Núñez-Álvarez, José Ricardo Sensors (Basel) Article Motor Imagery (MI)-based Brain–Computer Interfaces (BCIs) have been widely used as an alternative communication channel to patients with severe motor disabilities, achieving high classification accuracy through machine learning techniques. Recently, deep learning techniques have spotlighted the state-of-the-art of MI-based BCIs. These techniques still lack strategies to quantify predictive uncertainty and may produce overconfident predictions. In this work, methods to enhance the performance of existing MI-based BCIs are proposed in order to obtain a more reliable system for real application scenarios. First, the Monte Carlo dropout (MCD) method is proposed on MI deep neural models to improve classification and provide uncertainty estimation. This approach was implemented using Shallow Convolutional Neural Network (SCNN-MCD) and with an ensemble model (E-SCNN-MCD). As another contribution, to discriminate MI task predictions of high uncertainty, a threshold approach is introduced and tested for both SCNN-MCD and E-SCNN-MCD approaches. The BCI Competition IV Databases 2a and 2b were used to evaluate the proposed methods for both subject-specific and non-subject-specific strategies, obtaining encouraging results for MI recognition. MDPI 2021-10-30 /pmc/articles/PMC8588128/ /pubmed/34770553 http://dx.doi.org/10.3390/s21217241 Text en © 2021 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Milanés-Hermosilla, Daily
Trujillo Codorniú, Rafael
López-Baracaldo, René
Sagaró-Zamora, Roberto
Delisle-Rodriguez, Denis
Villarejo-Mayor, John Jairo
Núñez-Álvarez, José Ricardo
Monte Carlo Dropout for Uncertainty Estimation and Motor Imagery Classification
title Monte Carlo Dropout for Uncertainty Estimation and Motor Imagery Classification
title_full Monte Carlo Dropout for Uncertainty Estimation and Motor Imagery Classification
title_fullStr Monte Carlo Dropout for Uncertainty Estimation and Motor Imagery Classification
title_full_unstemmed Monte Carlo Dropout for Uncertainty Estimation and Motor Imagery Classification
title_short Monte Carlo Dropout for Uncertainty Estimation and Motor Imagery Classification
title_sort monte carlo dropout for uncertainty estimation and motor imagery classification
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8588128/
https://www.ncbi.nlm.nih.gov/pubmed/34770553
http://dx.doi.org/10.3390/s21217241
work_keys_str_mv AT milaneshermosilladaily montecarlodropoutforuncertaintyestimationandmotorimageryclassification
AT trujillocodorniurafael montecarlodropoutforuncertaintyestimationandmotorimageryclassification
AT lopezbaracaldorene montecarlodropoutforuncertaintyestimationandmotorimageryclassification
AT sagarozamoraroberto montecarlodropoutforuncertaintyestimationandmotorimageryclassification
AT delislerodriguezdenis montecarlodropoutforuncertaintyestimationandmotorimageryclassification
AT villarejomayorjohnjairo montecarlodropoutforuncertaintyestimationandmotorimageryclassification
AT nunezalvarezjosericardo montecarlodropoutforuncertaintyestimationandmotorimageryclassification