{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,30]],"date-time":"2026-07-30T09:03:52Z","timestamp":1785402232484,"version":"3.56.0"},"reference-count":41,"publisher":"MDPI AG","issue":"5","license":[{"start":{"date-parts":[[2026,4,28]],"date-time":"2026-04-28T00:00:00Z","timestamp":1777334400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"\u201cPioneer\u201d and \u201cLeading Goose\u201d R&D Program of Zhejiang","award":["2025C04001"],"award-info":[{"award-number":["2025C04001"]}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62571171"],"award-info":[{"award-number":["62571171"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"award":["62571171"],"award-info":[{"award-number":["62571171"]}],"id":[{"id":"https:\/\/ror.org\/01h0zpd94","id-type":"ROR","asserted-by":"publisher"}]},{"name":"MoE Humanities and Social Sciences Project","award":["24YJCZH225"],"award-info":[{"award-number":["24YJCZH225"]}]},{"DOI":"10.13039\/501100010255","name":"Guangxi Key Laboratory of Automatic Detecting Technology and Instruments","doi-asserted-by":"publisher","award":["YQ26204"],"award-info":[{"award-number":["YQ26204"]}],"id":[{"id":"10.13039\/501100010255","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Systems"],"abstract":"<jats:p>As a common brain-computer interface (BCI) paradigm, electroencephalogram (EEG)-based motor imagery provides a critical pathway for both assistive technology to (restoring communication and control) and active rehabilitation (promoting neural plasticity and functional recovery). Domain adaptation has been shown to effectively enhance the decoding performance of motor intentions for target subjects by leveraging labeled data from source subjects. However, EEG data from source subjects often contains extensive personal privacy, and the direct access to source EEG data easily leads to privacy leakage issues. An important research topic is to achieve domain adaptation without directly accessing the source subjects\u2019 raw data. To address this challenge, a privacy-preserving source-free domain adaptation framework, termed Transformer-based SFDA with Class-balanced Multicentric Dynamic Pseudo-labeling (T-CMDP), is proposed for cross-subject motor-imagery EEG classification. This framework consists of three coupled stages. In the source model training stage, a Transformer-based encoder combined with Riemannian manifold-aware feature extraction is employed to learn transferable and discriminative EEG feature representations. In the source-free target adaptation stage, only the pretrained source model is transferred to the target domain and adapted through knowledge distillation and information maximization, without accessing raw source EEG data. In the self-supervised learning stage, class-balanced multicentric prototypes and high-confidence pseudo-label updates are introduced to progressively refine the target-domain decision boundaries. Extensive experiments on three motor-imagery EEG datasets demonstrate that the proposed T-CMDP framework consistently outperforms eleven representative baselines from traditional machine learning, deep learning, and source-free transfer approaches, achieving average accuracies of 56.85%, 76.34%, and 74.49%, respectively. These results indicate that T-CMDP effectively alleviates inter-subject EEG distribution discrepancies and ensures the privacy preserving of source subjects, thereby facilitating more reliable and practical deployment of EEG-based BCI systems.<\/jats:p>","DOI":"10.3390\/systems14050476","type":"journal-article","created":{"date-parts":[[2026,4,28]],"date-time":"2026-04-28T15:09:33Z","timestamp":1777388973000},"page":"476","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Transformer-Based SFDA by Class-Balanced Multicentric Dynamic Pseudo-Labeling for Privacy-Preserving EEG-Based BCI Systems"],"prefix":"10.3390","volume":"14","author":[{"ORCID":"https:\/\/orcid.org\/0009-0008-9787-9782","authenticated-orcid":false,"given":"Jiangchuan","family":"Liu","sequence":"first","affiliation":[{"name":"HDU-ITMO Joint Institute, Hangzhou Dianzi University, Hangzhou 310018, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0007-0903-9701","authenticated-orcid":false,"given":"Jiatao","family":"Zhang","sequence":"additional","affiliation":[{"name":"HDU-ITMO Joint Institute, Hangzhou Dianzi University, Hangzhou 310018, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1846-2812","authenticated-orcid":false,"given":"Cong","family":"Hu","sequence":"additional","affiliation":[{"name":"Guangxi Key Laboratory of Automatic Detecting Technology and Instruments, Guilin University of Electronic Technology, Guilin 541004, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1208-972X","authenticated-orcid":false,"given":"Yong","family":"Peng","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology, Hangzhou Dianzi University, Hangzhou 310018, China"},{"name":"Key Laboratory of Brain-Machine Collaborative Intelligence of Zhejiang Province, Hangzhou 310018, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,4,28]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"1030","DOI":"10.1109\/TNSRE.2023.3236251","article-title":"Towards Practical BCI-Driven Wheelchairs: A Systematic Review Study","volume":"31","author":"Naser","year":"2023","journal-title":"IEEE Trans. Neural Syst. Rehabil. Eng."},{"key":"ref_2","doi-asserted-by":"crossref","unstructured":"Ukov, T., Sharabov, M., and Tsochev, G. (2025). A Systematic Review of the Need for Conceptual Models of Imagery Experiences. Systems, 13.","DOI":"10.3390\/systems13121051"},{"key":"ref_3","doi-asserted-by":"crossref","unstructured":"Y\u0131lmaz, S., Boz, C., \u00d6zsar\u0131, S.H., Y\u0131lmaz, F., Fidan T\u00fcrk\u00f6n, B., and Mete, A.H. (2025). Effects of Neurological Disorders on Health Expenditure and Economic Output: Dynamic Panel Analysis for OECD Countries. Systems, 13.","DOI":"10.3390\/systems13070521"},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"031002","DOI":"10.1088\/1741-2552\/ac74e0","article-title":"How to successfully classify EEG in motor imagery BCI: A metrological analysis of the state of the art","volume":"19","author":"Arpaia","year":"2022","journal-title":"J. Neural Eng."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1016\/j.neucom.2020.09.017","article-title":"A Review on Transfer Learning in EEG Signal Analysis","volume":"421","author":"Wan","year":"2021","journal-title":"Neurocomputing"},{"key":"ref_6","doi-asserted-by":"crossref","unstructured":"Li, W., and Peng, Y. (2022). Transfer EEG emotion recognition by combining semi-supervised regression with bipartite graph label propagation. Systems, 10.","DOI":"10.3390\/systems10040111"},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Yan, W., and Wu, Y. (2022). A Time-Frequency Denoising Method for Single-Channel Event-Related EEG. Front. Neurosci., 16.","DOI":"10.3389\/fnins.2022.991136"},{"key":"ref_8","first-page":"79","article-title":"EEG-Based Emotion Recognition: A Tutorial and Review","volume":"55","author":"Li","year":"2022","journal-title":"ACM Comput. Surv."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"639","DOI":"10.1109\/TPAMI.2014.2330598","article-title":"Probabilistic Common Spatial Patterns for Multichannel EEG Analysis","volume":"37","author":"Wu","year":"2015","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"1971","DOI":"10.1109\/JBHI.2024.3357995","article-title":"Manifold learning-based common spatial pattern for EEG signal classification","volume":"28","author":"Cai","year":"2024","journal-title":"IEEE J. Biomed. Health Inform."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"17","DOI":"10.1007\/s11571-021-09689-8","article-title":"A survey of brain network analysis by electroencephalographic signals","volume":"16","author":"Luo","year":"2022","journal-title":"Cogn. Neurodyn."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"14681","DOI":"10.1007\/s00521-021-06352-5","article-title":"Deep learning techniques for classification of electroencephalogram (EEG) motor imagery (MI) signals: A review","volume":"35","author":"Altaheri","year":"2023","journal-title":"Neural Comput. Appl."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"2767","DOI":"10.1109\/TNSRE.2023.3285309","article-title":"Global adaptive transformer for cross-subject enhanced EEG classification","volume":"31","author":"Song","year":"2023","journal-title":"IEEE Trans. Neural Syst. Rehabil. Eng."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"031004","DOI":"10.1088\/1741-2552\/ade290","article-title":"EEG-based affective brain-computer interfaces: Recent advancements and future challenges","volume":"22","author":"Chen","year":"2025","journal-title":"J. Neural Eng."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"120116","DOI":"10.1016\/j.neuroimage.2023.120116","article-title":"Stability, change, and reliable individual differences in electroencephalography measures: A lifespan perspective on progress and opportunities","volume":"275","author":"Lopez","year":"2023","journal-title":"NeuroImage"},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"102","DOI":"10.1186\/s40537-022-00652-w","article-title":"Transfer learning: A friendly introduction","volume":"9","author":"Hosna","year":"2022","journal-title":"J. Big Data"},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"104964","DOI":"10.1016\/j.cmpb.2019.06.023","article-title":"A transfer learning method with deep residual network for pediatric pneumonia diagnosis","volume":"187","author":"Liang","year":"2020","journal-title":"Comput. Methods Programs Biomed."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"107674","DOI":"10.1016\/j.patcog.2020.107674","article-title":"Attribute and Instance Weighted Naive Bayes","volume":"111","author":"Zhang","year":"2021","journal-title":"Pattern Recognit."},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"2312","DOI":"10.1109\/TCSS.2022.3184818","article-title":"Privacy-preserving brain\u2013computer interfaces: A systematic review","volume":"10","author":"Xia","year":"2023","journal-title":"IEEE Trans. Comput. Soc. Syst."},{"key":"ref_20","unstructured":"Kundu, J.N., Venkat, N., and Babu, R.V. (2020). Universal Source-Free Domain Adaptation. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, IEEE."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"77","DOI":"10.1017\/S0263574721000382","article-title":"Recent Approaches on Classification and Feature Extraction of EEG Signal: A Review","volume":"40","author":"Pooja","year":"2022","journal-title":"Robotica"},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Singh, A.K., and Krishnan, S. (2023). Trends in EEG Signal Feature Extraction Applications. Front. Artif. Intell., 5.","DOI":"10.3389\/frai.2022.1072801"},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"012032","DOI":"10.1088\/1757-899X\/557\/1\/012032","article-title":"PSD-Based Features Extraction for EEG Signal During Typing Task","volume":"Volume 557","author":"Ng","year":"2019","journal-title":"Proceedings of the IOP Conference Series: Materials Science and Engineering"},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"207","DOI":"10.1145\/3490237","article-title":"A Survey on Differential Privacy for Unstructured Data Content","volume":"54","author":"Zhao","year":"2022","journal-title":"ACM Comput. Surv. (CSUR)"},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"106854","DOI":"10.1016\/j.cie.2020.106854","article-title":"A Review of Applications in Federated Learning","volume":"149","author":"Li","year":"2020","journal-title":"Comput. Ind. Eng."},{"key":"ref_26","unstructured":"Liang, J., Hu, D., and Feng, J. (2020). Do We Really Need to Access the Source Data? Source Hypothesis Transfer for Unsupervised Domain Adaptation. Proceedings of the International Conference on Machine Learning, PMLR."},{"key":"ref_27","unstructured":"Niknam, Y. (2023). Source-Free Domain Adaptation for Sleep Stage Classification. [Master\u2019s Thesis, The University of Western Ontario (Canada)]."},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"2710","DOI":"10.1109\/TNSRE.2022.3207494","article-title":"Multi-Source Decentralized Transfer for Privacy-Preserving BCIs","volume":"30","author":"Zhang","year":"2022","journal-title":"IEEE Trans. Neural Syst. Rehabil. Eng."},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"3365","DOI":"10.1109\/TBME.2022.3168570","article-title":"Privacy-Preserving Domain Adaptation for Motor Imagery-Based Brain-Computer Interfaces","volume":"69","author":"Xia","year":"2022","journal-title":"IEEE Trans. Biomed. Eng."},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"938","DOI":"10.1109\/TCDS.2022.3193731","article-title":"Lightweight Source-Free Transfer for Privacy-Preserving Motor Imagery Classification","volume":"15","author":"Zhang","year":"2023","journal-title":"IEEE Trans. Cogn. Dev. Syst."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"2787","DOI":"10.1109\/TII.2023.3297323","article-title":"Source-free domain adaptation for privacy-preserving seizure prediction","volume":"20","author":"Zhao","year":"2024","journal-title":"IEEE Trans. Ind. Inform."},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Tangermann, M., M\u00fcller, K.R., Aertsen, A., Birbaumer, N., Braun, C., Brunner, C., Leeb, R., Mehring, C., Miller, K.J., and M\u00fcller-Putz, G.R. (2012). Review of the BCI Competition IV. Front. Neurosci., 6.","DOI":"10.3389\/fnins.2012.00055"},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"016003","DOI":"10.1088\/1741-2552\/ad171a","article-title":"Pseudo-online framework for BCI evaluation: A MOABB perspective using various MI and SSVEP datasets","volume":"21","author":"Carrara","year":"2024","journal-title":"J. Neural Eng."},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"17701","DOI":"10.1109\/TNNLS.2023.3307470","article-title":"Graph neural networks on spd manifolds for motor imagery classification: A perspective from the time\u2013frequency analysis","volume":"35","author":"Ju","year":"2024","journal-title":"IEEE Trans. Neural Netw. Learn. Syst."},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"399","DOI":"10.1109\/TBME.2019.2913914","article-title":"Transfer learning for brain\u2013computer interfaces: A Euclidean space data alignment approach","volume":"67","author":"He","year":"2020","journal-title":"IEEE Trans. Biomed. Eng."},{"key":"ref_36","doi-asserted-by":"crossref","first-page":"1117","DOI":"10.1109\/TNSRE.2020.2985996","article-title":"Manifold Embedded Knowledge Transfer for Brain-Computer Interfaces","volume":"28","author":"Zhang","year":"2020","journal-title":"IEEE Trans. Neural Syst. Rehabil. Eng."},{"key":"ref_37","doi-asserted-by":"crossref","unstructured":"Yang, H., Wu, F., Zhang, N., Chen, C., Wei, A., Peng, F., and Li, Z. (2023). A transfer learning method for motor imagery EEG signals classification based on CCSP and riemannian tangent space mapping. Proceedings of the 2023 IEEE International Conference on Real-Time Computing and Robotics (RCAR), IEEE.","DOI":"10.1109\/RCAR58764.2023.10249485"},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"1489","DOI":"10.1007\/s11571-023-09986-4","article-title":"Deep learning networks based decision fusion model of EEG and fNIRS for classification of cognitive tasks","volume":"18","author":"Rabbani","year":"2024","journal-title":"Cogn. Neurodyn."},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"139","DOI":"10.1145\/3422622","article-title":"Generative Adversarial Networks","volume":"63","author":"Goodfellow","year":"2020","journal-title":"Commun. ACM"},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"2484","DOI":"10.1109\/JBHI.2024.3431230","article-title":"EEG-DG: A Multi-Source Domain Generalization Framework for Motor Imagery EEG Classification","volume":"29","author":"Zhong","year":"2025","journal-title":"IEEE J. Biomed. Health Inform."},{"key":"ref_41","doi-asserted-by":"crossref","first-page":"16560","DOI":"10.1007\/s10489-022-04364-9","article-title":"Self-training transformer for source-free domain adaptation","volume":"53","author":"Yang","year":"2023","journal-title":"Appl. Intell."}],"container-title":["Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2079-8954\/14\/5\/476\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,28]],"date-time":"2026-04-28T15:43:48Z","timestamp":1777391028000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2079-8954\/14\/5\/476"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4,28]]},"references-count":41,"journal-issue":{"issue":"5","published-online":{"date-parts":[[2026,5]]}},"alternative-id":["systems14050476"],"URL":"https:\/\/doi.org\/10.3390\/systems14050476","relation":{},"ISSN":["2079-8954"],"issn-type":[{"value":"2079-8954","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4,28]]}}}