Academic Journal
Προβολή σε EDS

A Comprehensive Review of Sequence and Generative Models in Motor Imagery (MI) Classification for Brain-Computer Interfaces (BCIs).

Λεπτομέρειες βιβλιογραφικής εγγραφής
Τίτλος: A Comprehensive Review of Sequence and Generative Models in Motor Imagery (MI) Classification for Brain-Computer Interfaces (BCIs).
Συγγραφείς: Abbasi MA; School of Automation, Northwestern Polytechnical University, Xi'an 710072, China., Abbasi HF; School of Automation, Northwestern Polytechnical University, Xi'an 710072, China., Arsalan M; Department of Electrical Engineering, University of Engineering and Technology (UET), Lahore 54890, Pakistan., Khan D; College of Mechatronics and Control Engineering, Shenzhen University, Shenzhen 518060, China., Annuk A; Institute of Forestry and Engineering, Estonian University of Life Sciences, 51006 Tartu, Estonia., Yu X; School of Automation, Northwestern Polytechnical University, Xi'an 710072, China.
Πηγή: Sensors (Basel, Switzerland) [Sensors (Basel)] 2026 Aug 11; Vol. 26 (16). Date of Electronic Publication: 2026 Aug 11.
Τύπος έκδοσης: Journal Article; Review
Γλώσσα: English
Στοιχεία περιοδικού: Publisher: MDPI Country of Publication: Switzerland NLM ID: 101204366 Publication Model: Electronic Cited Medium: Internet ISSN: 1424-8220 (Electronic) Linking ISSN: 14248220 NLM ISO Abbreviation: Sensors (Basel) Subsets: MEDLINE
Imprint Name(s): Original Publication: Basel, Switzerland : MDPI, c2000-
Ιατρικοί όροι (MeSH): Imagination*/physiology , Brain-Computer Interfaces*, Brain/physiology ; Humans ; Generative Adversarial Networks ; Autoencoder ; Recurrent Neural Networks ; Signal Processing, Computer-Assisted ; Electroencephalography ; Neural Networks, Computer ; Generative Artificial Intelligence ; Deep Learning
Περίληψη: Motor imagery (MI) classification serves as the backbone to brain-computer interfaces (BCIs) by strengthening the communication bridge between the human brain and external peripheral devices. The past two decades have witnessed unprecedented success in MI-BCIs, with applications not only in medical fields but also in several other domains, such as gaming and robotic control. Initially, MI classification primarily relied on classical signal processing techniques that were heavily impacted by signal variations; however, recent trends in deep learning (DL), specifically in sequence-oriented, attention-based, hybrid, and generative architectures such as recurrent neural networks (RNNs), variational autoencoders (VAEs), generative adversarial networks (GANs), and transformers have significantly improved the efficiency and robustness of MI classification. This study presents a comprehensive review of these sequence-oriented, attention-based, hybrid, and generative architectures including RNNs, VAEs, GANs, and transformers, comparing their robustness across various public MI datasets, highlighting their challenges, such as inter-subject variation, low signal-to-noise ratio (SNR), and the obstacles in real-time signal classification. We perform an in-depth analysis on the strengths and limitations of traditional models such as RNNs and LSTMs as well as emergent models such as VAEs and transformers, which have demonstrated superior performance in extracting the intricate patterns of the EEG data with low latency. Moreover, we critically examine the future potential of such models in overcoming current bottlenecks, such as weak generalization on unseen data and high computational load. This study aims to assist researchers in attaining significant insights into the state-of-the-art sequence, attention-based, hybrid, and generative models used in MI classification, thus offering a direction for future innovation.
References: Comput Methods Programs Biomed. 2021 Jan;198:105808. (PMID: 33157470)
Front Hum Neurosci. 2024 Jun 05;18:1429130. (PMID: 38903409)
IEEE Trans Neural Syst Rehabil Eng. 2023;31:2767-2777. (PMID: 37307178)
J Neural Eng. 2018 Oct;15(5):056013. (PMID: 29932424)
IEEE Trans Neural Syst Rehabil Eng. 2023;31:646-656. (PMID: 37015688)
J Neurosci Methods. 2024 May;405:110108. (PMID: 38458260)
Brain Sci. 2023 Jan 12;13(1):. (PMID: 36672115)
Comput Methods Programs Biomed. 2021 Aug;207:106150. (PMID: 34034032)
Neuroscience. 2024 Sep 25;556:42-51. (PMID: 39103043)
IEEE Trans Neural Syst Rehabil Eng. 2022;30:540-549. (PMID: 35235515)
Conf Proc IEEE Eng Med Biol Soc. 2005;2005:5392-5. (PMID: 17281471)
Front Hum Neurosci. 2021 Jun 23;15:653659. (PMID: 34248521)
IEEE J Biomed Health Inform. 2023 Nov;27(11):5459-5470. (PMID: 37578918)
IEEE Trans Neural Syst Rehabil Eng. 2023;31:3132-3141. (PMID: 37478039)
IEEE Trans Neural Syst Rehabil Eng. 2025;33:1497-1508. (PMID: 40232894)
IEEE J Biomed Health Inform. 2024 Dec;28(12):7126-7137. (PMID: 39190517)
J Neural Eng. 2025 Jan 07;22(1):. (PMID: 39693767)
Comput Biol Med. 2022 Apr;143:105288. (PMID: 35168083)
Sci Rep. 2025 Apr 15;15(1):12935. (PMID: 40234486)
J Neurosci Methods. 2024 Aug;408:110182. (PMID: 38795979)
IEEE Trans Neural Syst Rehabil Eng. 2022;30:226-237. (PMID: 35041605)
J Neural Eng. 2024 May 17;21(3):. (PMID: 38757187)
IEEE Trans Biomed Eng. 2026 Jan 13;PP:. (PMID: 41528907)
J Neural Eng. 2023 Mar 03;20(2):. (PMID: 36763992)
J Neural Eng. 2024 Aug 14;21(4):. (PMID: 39116892)
Comput Biol Med. 2024 Jun;175:108504. (PMID: 38701593)
BMJ. 2021 Mar 29;372:n71. (PMID: 33782057)
IEEE J Biomed Health Inform. 2026 Feb;30(2):1153-1165. (PMID: 40627473)
Neural Netw. 2021 Dec;144:334-341. (PMID: 34547671)
Cogn Neurodyn. 2026 Dec;20(1):93. (PMID: 42153184)
IEEE Trans Neural Syst Rehabil Eng. 2023;31:710-719. (PMID: 37015413)
Front Comput Neurosci. 2023 Mar 02;17:1151852. (PMID: 36936191)
Comput Biol Med. 2023 Sep;164:107254. (PMID: 37499295)
J Neural Eng. 2022 Jun 17;19(3):. (PMID: 35640554)
Neural Netw. 2023 Aug;165:451-462. (PMID: 37336030)
IEEE Trans Neural Syst Rehabil Eng. 2023;31:863-874. (PMID: 37018635)
Neural Netw. 2024 Oct;178:106470. (PMID: 38943861)
Imaging Neurosci (Camb). 2025 Nov 05;3:. (PMID: 41211102)
Comput Biol Med. 2022 Sep;148:105931. (PMID: 35952544)
Sensors (Basel). 2025 May 05;25(9):. (PMID: 40363359)
J Neurosci Methods. 2023 Oct 1;398:109953. (PMID: 37611877)
J Neural Eng. 2024 Oct 07;:. (PMID: 39374625)
IEEE Rev Biomed Eng. 2025;18:26-49. (PMID: 39186407)
Heliyon. 2023 Feb 15;9(3):e13745. (PMID: 36851960)
Hum Brain Mapp. 2017 Nov;38(11):5391-5420. (PMID: 28782865)
IEEE Trans Neural Netw Learn Syst. 2021 Sep;32(9):4039-4051. (PMID: 32841127)
IEEE Trans Neural Netw Learn Syst. 2021 Feb;32(2):535-545. (PMID: 32745012)
IEEE Trans Neural Syst Rehabil Eng. 2022;30:2406-2417. (PMID: 35976835)
IEEE Trans Neural Syst Rehabil Eng. 2019 Jun;27(6):1200-1208. (PMID: 31095487)
J Neurosci Methods. 2023 Nov 1;399:109969. (PMID: 37683772)
IEEE J Biomed Health Inform. 2023 Aug;27(8):3818-3829. (PMID: 37200132)
IEEE Trans Neural Syst Rehabil Eng. 2021;29:2417-2424. (PMID: 34710045)
Neuroscience. 2023 Sep 1;527:64-73. (PMID: 37517788)
Sci Rep. 2026 Jan 20;16(1):2503. (PMID: 41559106)
IEEE Trans Neural Syst Rehabil Eng. 2022;30:2126-2136. (PMID: 35914032)
J Neural Eng. 2024 Jul 12;21(4):. (PMID: 38959876)
Bioengineering (Basel). 2026 Mar 06;13(3):. (PMID: 41899838)
Sci Rep. 2025 Jul 2;15(1):23380. (PMID: 40603471)
IEEE Trans Neural Syst Rehabil Eng. 2023;31:1137-1148. (PMID: 37022366)
J Neural Eng. 2021 Apr 06;18(4):046026. (PMID: 33821808)
Biomimetics (Basel). 2026 May 30;11(6):. (PMID: 42345666)
IEEE Trans Neural Syst Rehabil Eng. 2024;32:154-165. (PMID: 38090841)
IEEE Trans Neural Syst Rehabil Eng. 2024;32:1535-1545. (PMID: 38536681)
Grant Information: TARISTU24-TK12 Estonian Research Council
Contributed Indexing: Keywords: brain-computer interface (BCI); electroencephalography (EEG); generative adversarial networks (GANs); motor imagery (MI); recurrent neural networks (RNNs); transformers; variational autoencoders (VAEs)
Entry Date(s): Date Created: 20260827 Date Completed: 20260827 Latest Revision: 20260829
Update Code: 20260829
PubMed Central ID: PMC13517966
DOI: 10.3390/s26165097
PMID: 42655407
Βάση Δεδομένων: MEDLINE
Περιγραφή
ISSN:1424-8220
DOI:10.3390/s26165097