Brain–machine interface based on deep learning to control asynchronously a lower-limb robotic exoskeleton: a case-of-study
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, su...
Saved in:
| Published in | Journal of neuroengineering and rehabilitation Vol. 21; no. 1; pp. 48 - 14 |
|---|---|
| Main Authors | , , , , , , , |
| Format | Journal Article |
| Language | English |
| Published |
England
BioMed Central Ltd
05.04.2024
BioMed Central BMC |
| Subjects | |
| Online Access | Get full text |
Cover
Loading…
| Abstract | 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.
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.
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.
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. |
|---|---|
| AbstractList | 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. 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. 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. 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. 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.BACKGROUNDThis 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.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.METHODSA 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.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.RESULTSThe 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.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.CONCLUSIONThis 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. 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. Keywords: Brain-machine interface, EEG, Exoskeleton, Deep learning, Transfer learning Abstract 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. 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. 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. 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. 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. BackgroundThis 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.MethodsA 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.ResultsThe 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.ConclusionThis 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. |
| ArticleNumber | 48 |
| Audience | Academic |
| Author | Jones, Oscar Ferrero, Laura Navarro, Jacobo Ortiz, Mario Iáñez, Eduardo Soriano-Segura, Paula Contreras-Vidal, José L. Azorín, José M. |
| Author_xml | – sequence: 1 givenname: Laura orcidid: 0000-0003-2256-757X surname: Ferrero fullname: Ferrero, Laura – sequence: 2 givenname: Paula surname: Soriano-Segura fullname: Soriano-Segura, Paula – sequence: 3 givenname: Jacobo surname: Navarro fullname: Navarro, Jacobo – sequence: 4 givenname: Oscar surname: Jones fullname: Jones, Oscar – sequence: 5 givenname: Mario orcidid: 0000-0002-4269-1554 surname: Ortiz fullname: Ortiz, Mario – sequence: 6 givenname: Eduardo orcidid: 0000-0001-8057-5952 surname: Iáñez fullname: Iáñez, Eduardo – sequence: 7 givenname: José M. orcidid: 0000-0001-5548-9657 surname: Azorín fullname: Azorín, José M. – sequence: 8 givenname: José L. orcidid: 0000-0002-6499-1208 surname: Contreras-Vidal fullname: Contreras-Vidal, José L. |
| BackLink | https://www.ncbi.nlm.nih.gov/pubmed/38581031$$D View this record in MEDLINE/PubMed |
| BookMark | eNp9ksFu1DAQhiNURNuFF-CALHHhkmLHSRxzK1WBSpW4wNma2OOtF8de7KxgJQ68Q9-QJ8HbLQUqhCzL1uib356Z_7g6CDFgVT1l9ISxoX-ZWSOHtqZN2Yy3TS0fVEdMtLymlPKDP-6H1XHOq3Jpadc-qg750A2McnZUfXudwIUf368n0FcuIHFhxmRBIxkhoyExEIO4Jh4hBReWZI5ExzCn6AnkbdBXKYa4yX5LgPj4BVPt3TSSFMc4O03wa8yf0OMcw6tC6CJaR1vneWO2j6uHFnzGJ7fnovr45vzD2bv68v3bi7PTy1p3dJhryamQhhlpW4lCI0UhxchbLXSjJRu17anlIIWxHLUZ-m7XGDaCGRuQzPBFdbHXNRFWap3cBGmrIjh1E4hpqSCVz3pUaItM24l-oGMrWwtGaoEUEKHr-54XrRd7rXWKnzeYZzW5rNF7CFjaoHhpctNyJmlBn99DV3GTQql0R4lB8EH2v6kllPddsHFOoHei6lQMkkrelekuqpN_UGUZnFyZB1pX4n8lPLt9fDNOaO6q_jX6AjR7QKeYc0J7hzCqdv5Se3-p4i914y8lS9JwL0m7GWa3MwQ4_7_Un9l21JA |
| CitedBy_id | crossref_primary_10_3390_signals5030034 crossref_primary_10_1115_1_4066859 crossref_primary_10_1109_TBME_2024_3440036 crossref_primary_10_1109_TIM_2024_3481538 |
| Cites_doi | 10.1038/s41598-020-60932-4 10.1088/1741-2552/aace8c 10.1016/j.neuroscience.2016.11.023 10.1007/BF01129656 10.3390/mi13060927 10.1016/S1388-2457(99)00141-8 10.3389/fnhum.2018.00312 10.1007/978-3-642-15995-4_78 10.3389/fnins.2021.774857 10.3390/s20247309 10.1016/j.msksp.2020.102313 10.1016/j.nicl.2020.102502 10.1177/15459683221138751 10.1109/EMBC40787.2023.10340008 10.3389/fninf.2018.00078 10.3390/s21196431 10.1007/s11571-021-09676-z 10.1002/hbm.23730 10.1088/1741-2552/aab2f2 10.3389/fnins.2016.00456 10.1109/ACCESS.2020.2991812 10.1038/s41598-017-07823-3 10.1016/j.compbiomed.2022.105242 10.1016/j.isci.2023.106675 10.1152/jn.90989.2008 10.1109/EMBC44109.2020.9175929 10.1109/EMBC48229.2022.9871590 10.1016/j.jneumeth.2022.109736 10.1088/1741-2552/abda0c 10.1016/j.neuroimage.2005.12.003 10.3390/mi13091485 10.1123/jsep.34.5.621 10.3390/app11094106 10.1109/EMBC.2018.8512256 10.1109/EMBC46164.2021.9630155 10.1371/journal.pone.0268880 10.1088/1741-2552/aaa8c0 10.3389/fbioe.2020.00735 10.1088/1741-2552/ab0ab5 10.1109/EMBC.2019.8857575 10.1038/s41598-019-46310-9 10.1109/EMBC40787.2023.10340275 10.3390/s19061423 10.1088/1741-2560/13/2/026013 |
| ContentType | Journal Article |
| Copyright | 2024. The Author(s). COPYRIGHT 2024 BioMed Central Ltd. 2024. This work is licensed under http://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License. |
| Copyright_xml | – notice: 2024. The Author(s). – notice: COPYRIGHT 2024 BioMed Central Ltd. – notice: 2024. This work is licensed under http://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License. |
| DBID | AAYXX CITATION NPM 3V. 7QO 7RV 7TB 7TK 7TS 7X7 7XB 88C 88E 8FD 8FE 8FG 8FH 8FI 8FJ 8FK ABJCF ABUWG AFKRA AZQEC BBNVY BENPR BGLVJ BHPHI CCPQU DWQXO FR3 FYUFA GHDGH GNUQQ HCIFZ K9. KB0 L6V LK8 M0S M0T M1P M7P M7S NAPCQ P64 PHGZM PHGZT PIMPY PJZUB PKEHL PPXIY PQEST PQGLB PQQKQ PQUKI PTHSS 7X8 DOA |
| DOI | 10.1186/s12984-024-01342-9 |
| DatabaseName | CrossRef PubMed ProQuest Central (Corporate) Biotechnology Research Abstracts Nursing & Allied Health Database Mechanical & Transportation Engineering Abstracts Neurosciences Abstracts Physical Education Index ProQuest Health & Medical Collection ProQuest Central (purchase pre-March 2016) Healthcare Administration Database (Alumni) Medical Database (Alumni Edition) Technology Research Database ProQuest SciTech Collection ProQuest Technology Collection ProQuest Natural Science Collection Hospital Premium Collection Hospital Premium Collection (Alumni Edition) ProQuest Central (Alumni) (purchase pre-March 2016) Materials Science & Engineering Collection ProQuest Central (Alumni) ProQuest Central UK/Ireland ProQuest Central Essentials Biological Science Collection ProQuest Central Technology Collection Natural Science Collection ProQuest One Community College ProQuest Central Korea Engineering Research Database Health Research Premium Collection Health Research Premium Collection (Alumni) ProQuest Central Student SciTech Premium Collection ProQuest Health & Medical Complete (Alumni) Nursing & Allied Health Database (Alumni Edition) ProQuest Engineering Collection ProQuest Biological Science Collection Health & Medical Collection (Alumni Edition) Healthcare Administration Database (Proquest) Medical Database Biological Science Database Engineering Database Nursing & Allied Health Premium Biotechnology and BioEngineering Abstracts ProQuest Central Premium ProQuest One Academic Publicly Available Content Database ProQuest Health & Medical Research Collection ProQuest One Academic Middle East (New) ProQuest One Health & Nursing ProQuest One Academic Eastern Edition (DO NOT USE) ProQuest One Applied & Life Sciences ProQuest One Academic ProQuest One Academic UKI Edition Engineering Collection MEDLINE - Academic DOAJ Directory of Open Access Journals |
| DatabaseTitle | CrossRef PubMed Publicly Available Content Database ProQuest Central Student ProQuest Central Essentials SciTech Premium Collection ProQuest One Applied & Life Sciences Health Research Premium Collection Natural Science Collection Health & Medical Research Collection Biological Science Collection ProQuest Central (New) ProQuest Medical Library (Alumni) Engineering Collection Engineering Database ProQuest Biological Science Collection ProQuest One Academic Eastern Edition ProQuest Hospital Collection ProQuest Technology Collection Health Research Premium Collection (Alumni) Biological Science Database Neurosciences Abstracts ProQuest Hospital Collection (Alumni) Biotechnology and BioEngineering Abstracts Nursing & Allied Health Premium ProQuest Health & Medical Complete ProQuest One Academic UKI Edition ProQuest Health Management (Alumni Edition) ProQuest Nursing & Allied Health Source (Alumni) Engineering Research Database ProQuest One Academic ProQuest One Academic (New) Technology Collection Technology Research Database ProQuest One Academic Middle East (New) Mechanical & Transportation Engineering Abstracts ProQuest Health & Medical Complete (Alumni) ProQuest Central (Alumni Edition) ProQuest One Community College ProQuest One Health & Nursing ProQuest Natural Science Collection Physical Education Index ProQuest Central ProQuest Health & Medical Research Collection ProQuest Engineering Collection Biotechnology Research Abstracts Health and Medicine Complete (Alumni Edition) ProQuest Central Korea ProQuest Health Management ProQuest Nursing & Allied Health Source ProQuest SciTech Collection ProQuest Medical Library Materials Science & Engineering Collection ProQuest Central (Alumni) MEDLINE - Academic |
| DatabaseTitleList | MEDLINE - Academic PubMed Publicly Available Content Database |
| Database_xml | – sequence: 1 dbid: DOA name: DOAJ Directory of Open Access Journals url: https://www.doaj.org/ sourceTypes: Open Website – sequence: 2 dbid: NPM name: PubMed url: https://proxy.k.utb.cz/login?url=http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed sourceTypes: Index Database – sequence: 3 dbid: 8FG name: ProQuest Technology Collection url: https://search.proquest.com/technologycollection1 sourceTypes: Aggregation Database |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Medicine Engineering Occupational Therapy & Rehabilitation Physical Therapy |
| EISSN | 1743-0003 |
| EndPage | 14 |
| ExternalDocumentID | oai_doaj_org_article_ef97d457680b494fad9c7e0aeea56663 A789093501 38581031 10_1186_s12984_024_01342_9 |
| Genre | Journal Article |
| GeographicLocations | Spain Germany |
| GeographicLocations_xml | – name: Spain – name: Germany |
| GrantInformation_xml | – fundername: Ministry of Science, Innovation and Universities through the Aid for the Training of University Teachers grantid: FPU19/03165 – fundername: MCIN/AEI/10.13039/501100011033 and by ERDF A way of making Europe grantid: PID2021-124111OB-C31 |
| GroupedDBID | --- 0R~ 29L 2QV 2WC 53G 5GY 5VS 7RV 7X7 88E 8FE 8FG 8FH 8FI 8FJ AAFWJ AAJSJ AASML AAWTL AAYXX ABDBF ABJCF ABUWG ACGFO ACGFS ACIWK ACPRK ACUHS ADBBV ADRAZ ADUKV AENEX AFKRA AFPKN AFRAH AHBYD AHMBA AHYZX ALIPV ALMA_UNASSIGNED_HOLDINGS AMKLP AMTXH AOIJS AQUVI BAPOH BAWUL BBNVY BCNDV BENPR BFQNJ BGLVJ BHPHI BMC BPHCQ BVXVI C6C CCPQU CITATION CS3 DIK DU5 E3Z EBD EBLON EBS ESX F5P FYUFA GROUPED_DOAJ GX1 HCIFZ HMCUK HYE I-F IAO IHR INH INR IPY ITC KQ8 L6V LK8 M0T M1P M48 M7P M7S ML0 M~E NAPCQ O5R O5S OK1 OVT P2P PGMZT PHGZM PHGZT PIMPY PQQKQ PROAC PSQYO PTHSS RBZ RNS ROL RPM RSV SBL SOJ TR2 TUS UKHRP WOQ WOW XSB ~8M NPM PJZUB PPXIY PQGLB PMFND 3V. 7QO 7TB 7TK 7TS 7XB 8FD 8FK AZQEC DWQXO FR3 GNUQQ K9. P64 PKEHL PQEST PQUKI 7X8 PUEGO |
| ID | FETCH-LOGICAL-c508t-93079d1d9f49e7ce0e797b34c7c2c91bcf60f3a97df3ecd86512981badb2a91d3 |
| IEDL.DBID | M48 |
| ISSN | 1743-0003 |
| IngestDate | Wed Aug 27 01:30:13 EDT 2025 Fri Jul 11 08:16:23 EDT 2025 Fri Jul 25 19:12:16 EDT 2025 Tue Jun 17 22:10:11 EDT 2025 Tue Jun 10 21:08:29 EDT 2025 Mon Jul 21 05:50:48 EDT 2025 Thu Apr 24 23:12:32 EDT 2025 Tue Jul 01 02:20:02 EDT 2025 |
| IsDoiOpenAccess | true |
| IsOpenAccess | true |
| IsPeerReviewed | true |
| IsScholarly | true |
| Issue | 1 |
| Keywords | Deep learning Exoskeleton Brain–machine interface Transfer learning EEG |
| Language | English |
| License | 2024. The Author(s). |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-c508t-93079d1d9f49e7ce0e797b34c7c2c91bcf60f3a97df3ecd86512981badb2a91d3 |
| Notes | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14 content type line 23 |
| ORCID | 0000-0001-8057-5952 0000-0002-6499-1208 0000-0002-4269-1554 0000-0001-5548-9657 0000-0003-2256-757X |
| OpenAccessLink | https://www.proquest.com/docview/3037873896?pq-origsite=%requestingapplication% |
| PMID | 38581031 |
| PQID | 3037873896 |
| PQPubID | 55356 |
| PageCount | 14 |
| ParticipantIDs | doaj_primary_oai_doaj_org_article_ef97d457680b494fad9c7e0aeea56663 proquest_miscellaneous_3034243190 proquest_journals_3037873896 gale_infotracmisc_A789093501 gale_infotracacademiconefile_A789093501 pubmed_primary_38581031 crossref_primary_10_1186_s12984_024_01342_9 crossref_citationtrail_10_1186_s12984_024_01342_9 |
| ProviderPackageCode | CITATION AAYXX |
| PublicationCentury | 2000 |
| PublicationDate | 2024-04-05 |
| PublicationDateYYYYMMDD | 2024-04-05 |
| PublicationDate_xml | – month: 04 year: 2024 text: 2024-04-05 day: 05 |
| PublicationDecade | 2020 |
| PublicationPlace | England |
| PublicationPlace_xml | – name: England – name: London |
| PublicationTitle | Journal of neuroengineering and rehabilitation |
| PublicationTitleAlternate | J Neuroeng Rehabil |
| PublicationYear | 2024 |
| Publisher | BioMed Central Ltd BioMed Central BMC |
| Publisher_xml | – name: BioMed Central Ltd – name: BioMed Central – name: BMC |
| References | G Pfurtscheller (1342_CR11) 2006; 31 N Tibrewal (1342_CR39) 2022; 17 SY Gordleeva (1342_CR43) 2020; 8 K Xie (1342_CR3) 2017; 7 N Padfield (1342_CR7) 2019; 19 RT Schirrmeister (1342_CR18) 2017; 38 S Trapero-Asenjo (1342_CR29) 2021; 51 S Nakagome (1342_CR8) 2020; 10 Z Chang (1342_CR22) 2022; 13 P Wierzgała (1342_CR44) 2018; 12 1342_CR41 JJ Bird (1342_CR21) 2021; 18 P Chholak (1342_CR28) 2019; 9 1342_CR40 L Xu (1342_CR4) 2021; 15 1342_CR25 A Kilicarslan (1342_CR33) 2016; 13 J Zhang (1342_CR23) 2022; 13 A Gharabaghi (1342_CR1) 2016; 10 C Ruffino (1342_CR31) 2017; 341 L Ferrero (1342_CR36) 2021 M Ortiz (1342_CR14) 2020; 8 ZJ Koles (1342_CR15) 1990; 2 L Ferrero (1342_CR37) 2023; 26 A Colucci (1342_CR5) 2022; 36 M Tariq (1342_CR12) 2018; 12 F Lotte (1342_CR26) 2018; 15 SE Williams (1342_CR30) 2012; 34 1342_CR19 G Pfurtscheller (1342_CR10) 1999; 110 1342_CR13 1342_CR35 1342_CR16 1342_CR38 Y He (1342_CR6) 2018; 15 J Choi (1342_CR27) 2020; 20 MT Sadiq (1342_CR24) 2022; 143 NA Bhagat (1342_CR9) 2020; 28 KA Ludwig (1342_CR34) 2009; 101 A Craik (1342_CR17) 2019; 16 Z Khademi (1342_CR2) 2023; 383 P Barria (1342_CR42) 2021; 21 JS Huang (1342_CR20) 2021; 15 VJ Lawhern (1342_CR32) 2018; 15 |
| References_xml | – volume: 10 start-page: 4372 issue: 1 year: 2020 ident: 1342_CR8 publication-title: Sci Rep doi: 10.1038/s41598-020-60932-4 – volume: 15 start-page: 56013 issue: 5 year: 2018 ident: 1342_CR32 publication-title: J Neural Eng doi: 10.1088/1741-2552/aace8c – volume: 341 start-page: 61 year: 2017 ident: 1342_CR31 publication-title: Neuroscience doi: 10.1016/j.neuroscience.2016.11.023 – volume: 2 start-page: 275 issue: 4 year: 1990 ident: 1342_CR15 publication-title: Brain Topogr doi: 10.1007/BF01129656 – volume: 13 start-page: 927 issue: 6 year: 2022 ident: 1342_CR22 publication-title: Micromachines doi: 10.3390/mi13060927 – volume: 110 start-page: 1842 issue: 11 year: 1999 ident: 1342_CR10 publication-title: Clin Neurophysiol doi: 10.1016/S1388-2457(99)00141-8 – volume: 12 start-page: 312 issue: August year: 2018 ident: 1342_CR12 publication-title: Front Hum Neurosci doi: 10.3389/fnhum.2018.00312 – ident: 1342_CR16 doi: 10.1007/978-3-642-15995-4_78 – volume: 15 start-page: 1 issue: November year: 2021 ident: 1342_CR20 publication-title: Front Neurosci doi: 10.3389/fnins.2021.774857 – volume: 20 start-page: 7309 issue: 24 year: 2020 ident: 1342_CR27 publication-title: Sensors doi: 10.3390/s20247309 – volume: 51 issue: December 2020 year: 2021 ident: 1342_CR29 publication-title: Musculoskelet Sci Pract doi: 10.1016/j.msksp.2020.102313 – volume: 28 year: 2020 ident: 1342_CR9 publication-title: NeuroImage Clin doi: 10.1016/j.nicl.2020.102502 – volume: 36 start-page: 747 issue: 12 year: 2022 ident: 1342_CR5 publication-title: Neurorehabil Neural Repair doi: 10.1177/15459683221138751 – ident: 1342_CR35 doi: 10.1109/EMBC40787.2023.10340008 – volume: 12 start-page: 78 issue: 1 year: 2018 ident: 1342_CR44 publication-title: Front Neuroinform doi: 10.3389/fninf.2018.00078 – volume: 21 start-page: 6431 issue: 19 year: 2021 ident: 1342_CR42 publication-title: Sensors doi: 10.3390/s21196431 – volume: 15 start-page: 569 issue: 4 year: 2021 ident: 1342_CR4 publication-title: Cogn Neurodyn doi: 10.1007/s11571-021-09676-z – volume: 38 start-page: 5391 issue: 11 year: 2017 ident: 1342_CR18 publication-title: Hum Brain Mapp doi: 10.1002/hbm.23730 – volume: 15 issue: 3 year: 2018 ident: 1342_CR26 publication-title: J Neural Eng doi: 10.1088/1741-2552/aab2f2 – volume: 10 start-page: 456 year: 2016 ident: 1342_CR1 publication-title: Front Neurosci doi: 10.3389/fnins.2016.00456 – volume: 8 start-page: 84070 year: 2020 ident: 1342_CR43 publication-title: IEEE Access doi: 10.1109/ACCESS.2020.2991812 – volume: 7 start-page: 7808 issue: 1 year: 2017 ident: 1342_CR3 publication-title: Sci Rep doi: 10.1038/s41598-017-07823-3 – volume: 143 issue: August 2021 year: 2022 ident: 1342_CR24 publication-title: Comput Biol Med doi: 10.1016/j.compbiomed.2022.105242 – volume: 26 issue: 5 year: 2023 ident: 1342_CR37 publication-title: iScience doi: 10.1016/j.isci.2023.106675 – volume: 101 start-page: 1679 issue: 3 year: 2009 ident: 1342_CR34 publication-title: J Neurophysiol doi: 10.1152/jn.90989.2008 – ident: 1342_CR41 doi: 10.1109/EMBC44109.2020.9175929 – ident: 1342_CR38 doi: 10.1109/EMBC48229.2022.9871590 – volume: 383 issue: October 2022 year: 2023 ident: 1342_CR2 publication-title: J Neurosci Methods doi: 10.1016/j.jneumeth.2022.109736 – volume: 18 issue: 2 year: 2021 ident: 1342_CR21 publication-title: J Neural Eng doi: 10.1088/1741-2552/abda0c – volume: 31 start-page: 153 year: 2006 ident: 1342_CR11 publication-title: Neuroimage doi: 10.1016/j.neuroimage.2005.12.003 – volume: 13 start-page: 1 issue: 9 year: 2022 ident: 1342_CR23 publication-title: Micromachines doi: 10.3390/mi13091485 – volume: 34 start-page: 621 issue: 5 year: 2012 ident: 1342_CR30 publication-title: J Sport Exerc Psychol doi: 10.1123/jsep.34.5.621 – year: 2021 ident: 1342_CR36 publication-title: Appl Sci doi: 10.3390/app11094106 – ident: 1342_CR13 doi: 10.1109/EMBC.2018.8512256 – ident: 1342_CR19 doi: 10.1109/EMBC46164.2021.9630155 – volume: 17 start-page: 0268880 year: 2022 ident: 1342_CR39 publication-title: PLoS ONE doi: 10.1371/journal.pone.0268880 – volume: 15 issue: 2 year: 2018 ident: 1342_CR6 publication-title: J Neural Eng doi: 10.1088/1741-2552/aaa8c0 – volume: 8 start-page: 735 year: 2020 ident: 1342_CR14 publication-title: Front Bioeng Biotechnol doi: 10.3389/fbioe.2020.00735 – volume: 16 year: 2019 ident: 1342_CR17 publication-title: J Neural Eng doi: 10.1088/1741-2552/ab0ab5 – ident: 1342_CR25 doi: 10.1109/EMBC.2019.8857575 – volume: 9 start-page: 1 issue: 1 year: 2019 ident: 1342_CR28 publication-title: Sci Rep doi: 10.1038/s41598-019-46310-9 – ident: 1342_CR40 doi: 10.1109/EMBC40787.2023.10340275 – volume: 19 start-page: 1 issue: 6 year: 2019 ident: 1342_CR7 publication-title: Sensors (Switzerland) doi: 10.3390/s19061423 – volume: 13 issue: 2 year: 2016 ident: 1342_CR33 publication-title: J Neural Eng doi: 10.1088/1741-2560/13/2/026013 |
| SSID | ssj0034054 |
| Score | 2.4240363 |
| Snippet | 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... Background This research focused on the development of a motor imagery (MI) based brain-machine interface (BMI) using deep learning algorithms to control a... BackgroundThis research focused on the development of a motor imagery (MI) based brain–machine interface (BMI) using deep learning algorithms to control a... Abstract Background This research focused on the development of a motor imagery (MI) based brain–machine interface (BMI) using deep learning algorithms to... |
| SourceID | doaj proquest gale pubmed crossref |
| SourceType | Open Website Aggregation Database Index Database Enrichment Source |
| StartPage | 48 |
| SubjectTerms | Algorithms Analysis Brain Brain–machine interface Calibration Closed loops Cognitive tasks Decision making Deep learning EEG Electrodes Electroencephalography Exoskeleton Exoskeletons Experiments Feature extraction Feedback control Health aspects Machine learning Man-machine interfaces Mental task performance Methods Neural networks Questionnaires Rehabilitation Robot control Robotics Transfer learning User interface Wavelet transforms |
| SummonAdditionalLinks | – databaseName: DOAJ Directory of Open Access Journals dbid: DOA link: http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV1baxUxEA7SB9EHL8fbapUIXh4kdC_ZS3xrxVKEikgLfQu5TEQ83S1nT6GFPvgf_If-Emeyu4dzFPRF2KfdyZJkJjNfSOYbxl6U6PRkA42ooS6ENMHjmmsKUZS29Cm4OsS8tcOP1cGx_HBSnqyV-qI7YQM98DBxOxBU7SWh4tRKJYPxytWQGgCDSKSKPJ8Y86bN1OCDC4QhckqRaaqdHqNaIwXGI9w6FzIXaiMMRbb-P33yb0gzRpz9O-zWCBX57tDFu-watDN2c41AcMauH45H4zP2cp0vmB8NZAH8Ff-8QcU9Y7c_jZqZZO6xqz2qE_Hz-4_TeLMSOHFILIJxwCnIed613AOc8bHExBe-7Ph4x52b_rJ1xLDbnffzS274nOquifnXU8sXne2w8xwuuv4bxjfEmW9RwuFPRRdEpLa9z4733x-9OxBjVQbhEMwthUKvoHzmVZAKagcp1Kq2hXS1y53KrAtVGgqDWgsFON9UBCkQHBtvc6MyXzxgW23XwiPGUydtMCa36K6lq3Jjib2wtEUGxjZlk7BsUpJ24zxR5Yy5jluXptKDYjUqVkfFapWwN6s2ZwNhx1-l90j3K0ki244v0AT1aIL6XyaYsNdkOZpcAnbPmTGzAQdJ5Fp6l5KNFZ3gJmx7QxKXstv8PNmeHl1JrxFjoFNFXFkl7PnqM7Wk63EtoHZJRuYIBVWasIeDza6GRCe_VMvj8f8Y6hN2I4-rCJ9ym20tF-fwFFHZ0j6LC_AXL8404g priority: 102 providerName: Directory of Open Access Journals – databaseName: ProQuest Health & Medical Collection dbid: 7X7 link: http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwfV1ba9RAFB60guhD1dVqtMoIXh5kaJKZXMYXacVShIpIC_s2zLWI22TdbKEFH_wP_kN_iefMzq5dhUKekpOQybl9kznzHUJeVBD0ROtb1viGM6GDA59rOeOVqVzubRPivrXDT_XBsfg4rsbph9uQyiqXMTEGatdb_Ee-A6EWbAvSa_1u-p1h1yhcXU0tNK6TG0hdhiVdzXg14eIARsRyo0xb7wyQ21rBICvBBJqLksm1ZBQ5-_-PzP_gzZh39u-SzQQY6e5Cw_fINd-NyO1LNIIjcvMwLZCPyMvLrMH0aEEZQF_RL2uE3CNy53PSz1LmPvmxh90ifv_8dRrrKz1FJolZ0NZTTHWO9h113k9pajRxQuc9TZXuVA8XnUWe3f5smFxQTSfYfY1Nvp4aOutNDy9P_Xk_fIMsB2jzLUhYeCjrA4sEtw_I8f6Ho_cHLPVmYBYg3ZxJiA3SFU4GIX1jfe4b2RgubGNLKwtjQ50HrmXjAvfWtTUCC4DI2plSy8LxLbLR9Z1_RGhuhQlalwaCtrB1qQ1yGFaGF16btmozUiyVpGz6Ttg_Y6LiBKat1UKxChSromKVzMib1T3TBW3HldJ7qPuVJFJuxxP97EQlD1Y-wGAETs9yI6QI2knb-Fx7rwES1zwjr9FyFAYGeD2r0_4GGCRSbKld3HIscR03I9trkuDQdv3y0vZUCiiD-mv-GXm-uox3YpFc50G7KCNKAIQyz8jDhc2uhoTrv9jR4_HVD39CbpXRP-CotsnGfHbmnwLqmptn0bX-APBuK9c priority: 102 providerName: ProQuest |
| Title | Brain–machine interface based on deep learning to control asynchronously a lower-limb robotic exoskeleton: a case-of-study |
| URI | https://www.ncbi.nlm.nih.gov/pubmed/38581031 https://www.proquest.com/docview/3037873896 https://www.proquest.com/docview/3034243190 https://doaj.org/article/ef97d457680b494fad9c7e0aeea56663 |
| Volume | 21 |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwjV3ra9swEBd9wNg-7JG9vHVBgz0-DG2OJVvWYIxmNCuDlFIayDchyXIZS-0uSaHZX7872Q7NVsbA-IN9MpZPd_eTpfsdIa9ScHoi9zmTXnImTFmAzeWc8dSmReydLEPe2vgoO5yIb9N0ukW6ckftB1zcOLXDelKT-ez91c_VZzD4T8Hg8-zDAmJWLhhEG5gYc5EwtU12ITJJNNSxWK8qcAAnokmQ5JhNzbskmhufsRGoAp__3177DywaYtLoPrnbgkm632j_AdnyVY_cuUYx2CO3xu3ieY-8vs4oTE8bOgH6hp5skHX3yL3jVnedzEPya4iVJNh52HnpKXJMzEvjPMUgWNC6ooX3F7QtQXFGlzVt98BTs1hVDhl468vFbEUNnWFdNjb7fm7pvLY1vDr1V_XiB8Q_wKEfQcLBQ1ldskB9-4hMRgenXw5ZW7WBOQB7S6bAa6hiUKhSKC-dj71U0nLhpEucGlhXZnHJjZJFyb0r8gwhB4BnU9jEqEHBH5Odqq78U0JjJ2xpTGLBnQuXJcYiu2Fq-cAbm6d5RAadirRrvxJW1pjpMLXJM92oVYNadVCrVhF5t25z0RB6_FN6iJpfSyIZd7hQz890a9val9AZgRO32AolSlMoJ31svDcAljMekbc4bjQOYng9Z9rMB-gkkm_pfUxGVrjCG5G9DUkwdbd5uxt5urMUDRgEnC7gziwiL9e3sSVun6s8aBdlRAJQUcURedKM2HWXcGUYa308-4_Wz8ntJJgIHOke2VnOL_0LAGVL2yfbcirhnI--9snu8ODo-KQffnD0gw3-BjP_NUg |
| linkProvider | Scholars Portal |
| linkToHtml | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV3bbhMxEB2VInF54BJuCwWMROkDsrrZ9V6MhFALVCltKoRSKW_G9norRLobklQQiQf-gf_go_gSZvYSGpD6VilP2bG1zlzOTGyfAXgaYdATqUt54pKQC51n6HNpyMPIRJnvbJJX99b6B3HvULwbRsMV-NXehaFjlW1MrAJ1Vlr6j3wTQy3aFsJr_Gr8hVPXKNpdbVto1Gax5-ZfsWSbvtx9g_pdD4Kdt4PXPd50FeAWk5EZl2jVMutmMhfSJdb5LpGJCYVNbGBl19g89vNQyyTLQ2ezNCZIxOROZybQspuFOO8FuIjA65NHJcNFgRdi8iPaizlpvDmlgYIjCmLBHoqAyyXwq3oE_I8E_-S3Fc7t3IBrTYLKtmqLugkrrujA1VO0hR241G825DuwfpqlmA1qigL2jH1YIgDvwPX3jT20Mrfg-zZ1p_j94-dxdZ7TMWKumOTaOkbQmrGyYJlzY9Y0tjhis5I1J-uZns4LS7y-5cl0NGeajajbGx99OjZsUpoSX565b-X0M6IqZrcvUMLipLzMeUWoexsOz0Vrd2C1KAt3D5hvhcm1DgyChLBxoA1xJkYm7Dpt0ij1oNsqSdnmd6J-HSNVFUxprGrFKlSsqhSrpAfPF2PGNU3ImdLbpPuFJFF8V1-UkyPVRAzlclyMoHLQN0KKXGfSJs7XzmlMwePQgw2yHEWBCF_P6uY-BS6SKL3UFl1xlrRv7MHakiQGELv8uLU91QSwqfrrbh48WTymkXQor3CoXZIRAfqB9D24W9vsYkm030wdRO6fPfljuNwb9PfV_u7B3gO4ElS-gp9oDVZnkxP3EDO-mXlUuRmDj-ft138ALq1pxg |
| openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Brain-machine+interface+based+on+deep+learning+to+control+asynchronously+a+lower-limb+robotic+exoskeleton%3A+a+case-of-study&rft.jtitle=Journal+of+neuroengineering+and+rehabilitation&rft.au=Ferrero%2C+Laura&rft.au=Soriano-Segura%2C+Paula&rft.au=Navarro%2C+Jacobo&rft.au=Jones%2C+Oscar&rft.date=2024-04-05&rft.issn=1743-0003&rft.eissn=1743-0003&rft.volume=21&rft.issue=1&rft.spage=48&rft_id=info:doi/10.1186%2Fs12984-024-01342-9&rft.externalDBID=NO_FULL_TEXT |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1743-0003&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1743-0003&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1743-0003&client=summon |