Brain–machine interface based on deep learning to control asynchronously a lower-limb robotic exoskeleton: a case-of-study
Background This research focused on the development of a motor imagery (MI) based brain–machine interface (BMI) using deep learning algorithms to control a lower-limb robotic exoskeleton. The study aimed to overcome the limitations of traditional BMI approaches by leveraging the advantages of deep l...
| Autores: | , , , , , , , |
|---|---|
| Tipo de recurso: | artículo |
| Fecha de publicación: | 2024 |
| País: | España |
| Institución: | Universidad Miguel Hernández de Elche |
| Repositorio: | REDIUMH. Depósito Digital de la UMH |
| OAI Identifier: | oai:dnet:rediumh_____::7688aea7266879683a946cb880a239db |
| Acceso en línea: | https://hdl.handle.net/11000/39639 |
| Access Level: | acceso abierto |
| Palabra clave: | brain–machine interface EEG exoskeleton deep learning transfer learning CDU::6 - Ciencias aplicadas::62 - Ingeniería. Tecnología CDU::6 - Ciencias aplicadas::61 - Medicina::612 - Fisiología CDU::0 - Generalidades.::04 - Ciencia y tecnología de los ordenadores. Informática. |
| id |
ES_ae95cc2ac9462ef7ebcccec8f0ae8857 |
|---|---|
| oai_identifier_str |
oai:dnet:rediumh_____::7688aea7266879683a946cb880a239db |
| network_acronym_str |
ES |
| network_name_str |
España |
| repository_id_str |
|
| spelling |
Brain–machine interface based on deep learning to control asynchronously a lower-limb robotic exoskeleton: a case-of-studyFerrero, LauraSoriano Segura, PaulaNavarro, JacoboJones, OscarOrtiz, MarioIáñez, EduardoAzorín, José M.Contreras Vidal, José L.brain–machine interfaceEEGexoskeletondeep learningtransfer learningCDU::6 - Ciencias aplicadas::62 - Ingeniería. TecnologíaCDU::6 - Ciencias aplicadas::61 - Medicina::612 - FisiologíaCDU::0 - Generalidades.::04 - Ciencia y tecnología de los ordenadores. Informática.Background This research focused on the development of a motor imagery (MI) based brain–machine interface (BMI) using deep learning algorithms to control a lower-limb robotic exoskeleton. The study aimed to overcome the limitations of traditional BMI approaches by leveraging the advantages of deep learning, such as automated feature extraction and transfer learning. The experimental protocol to evaluate the BMI was designed as asynchronous, allowing subjects to perform mental tasks at their own will. Methods A total of five healthy able-bodied subjects were enrolled in this study to participate in a series of experimental sessions. The brain signals from two of these sessions were used to develop a generic deep learning model through transfer learning. Subsequently, this model was fine-tuned during the remaining sessions and subjected to evaluation. Three distinct deep learning approaches were compared: one that did not undergo fine-tuning, another that fine-tuned all layers of the model, and a third one that fine-tuned only the last three layers. The evaluation phase involved the exclusive closed-loop control of the exoskeleton device by the participants’ neural activity using the second deep learning approach for the decoding. Results The three deep learning approaches were assessed in comparison to an approach based on spatial features that was trained for each subject and experimental session, demonstrating their superior performance. Interestingly, the deep learning approach without fine-tuning achieved comparable performance to the features-based approach, indicating that a generic model trained on data from different individuals and previous sessions can yield similar efficacy. Among the three deep learning approaches compared, fine-tuning all layer weights demonstrated the highest performance. Conclusion This research represents an initial stride toward future calibration-free methods. Despite the efforts to diminish calibration time by leveraging data from other subjects, complete elimination proved unattainable. The study’s discoveries hold notable significance for advancing calibration-free approaches, offering the promise of minimizing the need for training trials. Furthermore, the experimental evaluation protocol employed in this study aimed to replicate real-life scenarios, granting participants a higher degree of autonomy in decision-making regarding actions such as walking or stopping gait.BioMed CentralDepartamentos de la UMH::Ingeniería Mecánica y Energía202620262024info:eu-repo/semantics/articleapplication/pdf14application/pdfhttps://hdl.handle.net/11000/39639reponame:REDIUMH. Depósito Digital de la UMHinstname:Universidad Miguel Hernández de ElcheInglésVol. 21Nº 48https://doi.org/10.1186/s12984-024-01342-9info:eu-repo/semantics/openAccessAttribution-NonCommercial-NoDerivatives 4.0 Internacionalhttp://creativecommons.org/licenses/by-nc-nd/4.0/oai:dnet:rediumh_____::7688aea7266879683a946cb880a239db2026-05-27T13:36:21Z |
| dc.title.none.fl_str_mv |
Brain–machine interface based on deep learning to control asynchronously a lower-limb robotic exoskeleton: a case-of-study |
| title |
Brain–machine interface based on deep learning to control asynchronously a lower-limb robotic exoskeleton: a case-of-study |
| spellingShingle |
Brain–machine interface based on deep learning to control asynchronously a lower-limb robotic exoskeleton: a case-of-study Ferrero, Laura brain–machine interface EEG exoskeleton deep learning transfer learning CDU::6 - Ciencias aplicadas::62 - Ingeniería. Tecnología CDU::6 - Ciencias aplicadas::61 - Medicina::612 - Fisiología CDU::0 - Generalidades.::04 - Ciencia y tecnología de los ordenadores. Informática. |
| title_short |
Brain–machine interface based on deep learning to control asynchronously a lower-limb robotic exoskeleton: a case-of-study |
| title_full |
Brain–machine interface based on deep learning to control asynchronously a lower-limb robotic exoskeleton: a case-of-study |
| title_fullStr |
Brain–machine interface based on deep learning to control asynchronously a lower-limb robotic exoskeleton: a case-of-study |
| title_full_unstemmed |
Brain–machine interface based on deep learning to control asynchronously a lower-limb robotic exoskeleton: a case-of-study |
| title_sort |
Brain–machine interface based on deep learning to control asynchronously a lower-limb robotic exoskeleton: a case-of-study |
| dc.creator.none.fl_str_mv |
Ferrero, Laura Soriano Segura, Paula Navarro, Jacobo Jones, Oscar Ortiz, Mario Iáñez, Eduardo Azorín, José M. Contreras Vidal, José L. |
| author |
Ferrero, Laura |
| author_facet |
Ferrero, Laura Soriano Segura, Paula Navarro, Jacobo Jones, Oscar Ortiz, Mario Iáñez, Eduardo Azorín, José M. Contreras Vidal, José L. |
| author_role |
author |
| author2 |
Soriano Segura, Paula Navarro, Jacobo Jones, Oscar Ortiz, Mario Iáñez, Eduardo Azorín, José M. Contreras Vidal, José L. |
| author2_role |
author author author author author author author |
| dc.contributor.none.fl_str_mv |
Departamentos de la UMH::Ingeniería Mecánica y Energía |
| dc.subject.none.fl_str_mv |
brain–machine interface EEG exoskeleton deep learning transfer learning CDU::6 - Ciencias aplicadas::62 - Ingeniería. Tecnología CDU::6 - Ciencias aplicadas::61 - Medicina::612 - Fisiología CDU::0 - Generalidades.::04 - Ciencia y tecnología de los ordenadores. Informática. |
| topic |
brain–machine interface EEG exoskeleton deep learning transfer learning CDU::6 - Ciencias aplicadas::62 - Ingeniería. Tecnología CDU::6 - Ciencias aplicadas::61 - Medicina::612 - Fisiología CDU::0 - Generalidades.::04 - Ciencia y tecnología de los ordenadores. Informática. |
| description |
Background This research focused on the development of a motor imagery (MI) based brain–machine interface (BMI) using deep learning algorithms to control a lower-limb robotic exoskeleton. The study aimed to overcome the limitations of traditional BMI approaches by leveraging the advantages of deep learning, such as automated feature extraction and transfer learning. The experimental protocol to evaluate the BMI was designed as asynchronous, allowing subjects to perform mental tasks at their own will. Methods A total of five healthy able-bodied subjects were enrolled in this study to participate in a series of experimental sessions. The brain signals from two of these sessions were used to develop a generic deep learning model through transfer learning. Subsequently, this model was fine-tuned during the remaining sessions and subjected to evaluation. Three distinct deep learning approaches were compared: one that did not undergo fine-tuning, another that fine-tuned all layers of the model, and a third one that fine-tuned only the last three layers. The evaluation phase involved the exclusive closed-loop control of the exoskeleton device by the participants’ neural activity using the second deep learning approach for the decoding. Results The three deep learning approaches were assessed in comparison to an approach based on spatial features that was trained for each subject and experimental session, demonstrating their superior performance. Interestingly, the deep learning approach without fine-tuning achieved comparable performance to the features-based approach, indicating that a generic model trained on data from different individuals and previous sessions can yield similar efficacy. Among the three deep learning approaches compared, fine-tuning all layer weights demonstrated the highest performance. Conclusion This research represents an initial stride toward future calibration-free methods. Despite the efforts to diminish calibration time by leveraging data from other subjects, complete elimination proved unattainable. The study’s discoveries hold notable significance for advancing calibration-free approaches, offering the promise of minimizing the need for training trials. Furthermore, the experimental evaluation protocol employed in this study aimed to replicate real-life scenarios, granting participants a higher degree of autonomy in decision-making regarding actions such as walking or stopping gait. |
| publishDate |
2024 |
| dc.date.none.fl_str_mv |
2024 2026 2026 |
| dc.type.none.fl_str_mv |
info:eu-repo/semantics/article |
| format |
article |
| dc.identifier.none.fl_str_mv |
https://hdl.handle.net/11000/39639 |
| url |
https://hdl.handle.net/11000/39639 |
| dc.language.none.fl_str_mv |
Inglés |
| language_invalid_str_mv |
Inglés |
| dc.relation.none.fl_str_mv |
Vol. 21 Nº 48 https://doi.org/10.1186/s12984-024-01342-9 |
| dc.rights.none.fl_str_mv |
info:eu-repo/semantics/openAccess Attribution-NonCommercial-NoDerivatives 4.0 Internacional http://creativecommons.org/licenses/by-nc-nd/4.0/ |
| eu_rights_str_mv |
openAccess |
| rights_invalid_str_mv |
Attribution-NonCommercial-NoDerivatives 4.0 Internacional http://creativecommons.org/licenses/by-nc-nd/4.0/ |
| dc.format.none.fl_str_mv |
application/pdf 14 application/pdf |
| dc.publisher.none.fl_str_mv |
BioMed Central |
| publisher.none.fl_str_mv |
BioMed Central |
| dc.source.none.fl_str_mv |
reponame:REDIUMH. Depósito Digital de la UMH instname:Universidad Miguel Hernández de Elche |
| instname_str |
Universidad Miguel Hernández de Elche |
| reponame_str |
REDIUMH. Depósito Digital de la UMH |
| collection |
REDIUMH. Depósito Digital de la UMH |
| repository.name.fl_str_mv |
|
| repository.mail.fl_str_mv |
|
| _version_ |
1869416593060003840 |
| score |
15.81155 |