{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,8]],"date-time":"2026-07-08T10:17:14Z","timestamp":1783505834707,"version":"3.55.0"},"reference-count":27,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2026,7,8]],"date-time":"2026-07-08T00:00:00Z","timestamp":1783468800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,7,8]],"date-time":"2026-07-08T00:00:00Z","timestamp":1783468800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100001784","name":"Victoria University","doi-asserted-by":"crossref","id":[{"id":"10.13039\/501100001784","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Health Inf Sci Syst"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Purpose<\/jats:title>\n                    <jats:p>The rise of large language models (LLMs) such as GPT-4 and DeepSeek has transformed healthcare information processing by enabling natural language-based clinical reasoning. However, the integration of LLMs with privacy-sensitive biomedical signals, particularly electroencephalogram (EEG) data used in brain-computer interface (BCI) systems, remains underexplored. EEG signals, especially during motor imagery (MI) tasks, are critical for assistive neurotechnologies but pose significant privacy risks due to their capacity to reveal cognitive and medical information. Traditional encryption techniques often distort signal structure or require decryption with additional noise, compromising classification performance and real-time usability.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>To address this gap, we propose a deep denoising structure-preserving neural encoding network (DSNet) that enables accurate classification of privacy-preserving encoded EEG representations without requiring decryption. EEG features were extracted using common spatial pattern (CSP) and transformed into privacy-preserving encoded representations while preserving their statistical structure. Here, encoding refers to a non-reversible neural transformation designed for privacy preservation rather than a formal cryptographic guarantee. Two deep learning architectures, a feedforward neural network (NN) and a recurrent neural network (RNN), were evaluated for classification in the encoded feature space. Furthermore, we integrated an LLM (GPT-4) to generate clinical-style summaries based on model outputs, enhancing interpretability for clinician review and potential clinical support use.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results and conclusion<\/jats:title>\n                    <jats:p>Using publicly available datasets, DSNet-NN achieved over 87% accuracy for every subject, outperforming both the RNN variant and baseline models. It also demonstrated resilience to simulated privacy attacks. LLM-generated reports provided clinician-friendly interpretations of MI predictions, supporting potential real-world applicability. This study introduces an AI framework that bridges privacy-preserving EEG decoding with LLM-based clinical reasoning, offering a practical solution for privacy-preserving neurorehabilitation and digital health systems.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1007\/s13755-026-00470-x","type":"journal-article","created":{"date-parts":[[2026,7,8]],"date-time":"2026-07-08T09:42:59Z","timestamp":1783503779000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["A novel deep learning approach for privacy-preserving encoded EEG-based brain-computer interfaces with clinical LLM applications"],"prefix":"10.1007","volume":"14","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-5842-930X","authenticated-orcid":false,"given":"Taslima","family":"Khanam","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Siuly","family":"Siuly","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kate","family":"Wang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hua","family":"Wang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,7,8]]},"reference":[{"issue":"2","key":"470_CR1","doi-asserted-by":"publisher","first-page":"827","DOI":"10.1007\/s11280-022-01076-5","volume":"26","author":"M You","year":"2023","unstructured":"You M, Yin J, Wang H, Cao J, Wang K, Miao Y, et al. A knowledge graph empowered online learning framework for access control decision-making. World Wide Web. 2023;26(2):827\u201348.","journal-title":"World Wide Web"},{"issue":"1","key":"470_CR2","doi-asserted-by":"publisher","first-page":"82","DOI":"10.1016\/j.jneumeth.2007.09.022","volume":"167","author":"K-R M\u00fcller","year":"2008","unstructured":"M\u00fcller K-R, Tangermann M, Dornhege G, Krauledat M, Curio G, Blankertz B. Machine learning for real-time single-trial EEG-analysis: from brain-computer interfacing to mental state monitoring. J Neurosci Methods. 2008;167(1):82\u201390.","journal-title":"J Neurosci Methods"},{"key":"470_CR3","unstructured":"Mane R, Chew E, Chua K, Ang KK, Robinson N, Vinod AP, et al. Fbcnet: a multi-view convolutional neural network for brain-computer interface. 2021.arXiv:2104.01233 arXiv preprint."},{"key":"470_CR4","doi-asserted-by":"publisher","first-page":"105242","DOI":"10.1016\/j.compbiomed.2022.105242","volume":"143","author":"MT Sadiq","year":"2022","unstructured":"Sadiq MT, Aziz MZ, Almogren A, Yousaf A, Siuly S, Rehman AU. Exploiting pretrained CNN models for the development of an EEG-based robust BCI framework. Comput Biol Med. 2022;143:105242.","journal-title":"Comput Biol Med"},{"key":"470_CR5","first-page":"103","volume-title":"Australasian Database Conference","author":"T Khanam","year":"2022","unstructured":"Khanam T, Siuly S, Wang H. Analysing big brain signal data for advanced brain computer interface system. In: Australasian Database Conference. Springer; 2022. p. 103\u201314."},{"issue":"4","key":"470_CR6","doi-asserted-by":"publisher","first-page":"799","DOI":"10.1007\/s00521-014-1753-3","volume":"26","author":"S Siuly","year":"2015","unstructured":"Siuly S, Li Y. Discriminating the brain activities for brain-computer interface applications through the optimal allocation-based approach. Neural Comput Appl. 2015;26(4):799\u2013811.","journal-title":"Neural Comput Appl"},{"issue":"9","key":"470_CR7","doi-asserted-by":"publisher","first-page":"6623","DOI":"10.1007\/s00521-022-08027-1","volume":"35","author":"T Khanam","year":"2023","unstructured":"Khanam T, Siuly S, Wang H. An optimized artificial intelligence based technique for identifying motor imagery from EEGs for advanced brain computer interface technology. Neural Comput Appl. 2023;35(9):6623\u201334.","journal-title":"Neural Comput Appl"},{"issue":"10","key":"470_CR8","doi-asserted-by":"publisher","first-page":"0335511","DOI":"10.1371\/journal.pone.0335511","volume":"20","author":"T Khanam","year":"2025","unstructured":"Khanam T, Siuly S, Ahmad K, Wang H. A novel channel reduction concept to enhance the classification of motor imagery tasks in brain-computer interface systems. PLoS ONE. 2025;20(10):0335511.","journal-title":"PLoS ONE"},{"key":"470_CR9","first-page":"209","volume-title":"International Conference on Health Information Science","author":"T Khanam","year":"2024","unstructured":"Khanam T, Siuly S, Wang K, Wang H. An ai driven framework for EEG based-BCI technology. In: International Conference on Health Information Science. Springer; 2024. p. 209\u201320."},{"issue":"2","key":"470_CR10","doi-asserted-by":"publisher","first-page":"938","DOI":"10.1109\/TCDS.2022.3193731","volume":"15","author":"W Zhang","year":"2022","unstructured":"Zhang W, Wu D. Lightweight source-free transfer for privacy-preserving motor imagery classification. IEEE Trans Cogn Dev Syst. 2022;15(2):938\u201349.","journal-title":"IEEE Trans Cogn Dev Syst"},{"key":"470_CR11","unstructured":"Martinovic I, Davies D, Frank M, Perito D, Ros T, Song D. On the feasibility of $$\\{$$Side-Channel$$\\}$$ attacks with $$\\{$$Brain-Computer$$\\}$$ interfaces. In: 21st USENIX Security Symposium (USENIX Security 12), 2012. p. 143\u201358"},{"key":"470_CR12","doi-asserted-by":"crossref","unstructured":"Peng, Y, Liu J, Liu H, Padfield N, Li J, Kong W, Lu B-L, Cichocki A. Prediction consistency and confidence-based proxy domain construction for privacy-preserving in cross-subject EEG classification. IEEE J Biomed Health Inf. 2025","DOI":"10.1109\/JBHI.2025.3595826"},{"issue":"8","key":"470_CR13","doi-asserted-by":"publisher","first-page":"1546","DOI":"10.1109\/TNSRE.2019.2926965","volume":"27","author":"A Agarwal","year":"2019","unstructured":"Agarwal A, Dowsley R, McKinney ND, Wu D, Lin C-T, De Cock M, et al. Protecting privacy of users in brain-computer interface applications. IEEE Trans Neural Syst Rehabil Eng. 2019;27(8):1546\u201355.","journal-title":"IEEE Trans Neural Syst Rehabil Eng"},{"key":"470_CR14","first-page":"102567","volume":"54","author":"Y Liu","year":"2020","unstructured":"Liu Y, Huang H, Xiao F, Malekian R, Wang W. Classification and recognition of encrypted EEG data based on neural network. J Inf Secur Appl. 2020;54:102567.","journal-title":"J Inf Secur Appl"},{"issue":"2","key":"470_CR15","doi-asserted-by":"publisher","first-page":"2520","DOI":"10.1109\/JIOT.2023.3292232","volume":"11","author":"N Yan","year":"2023","unstructured":"Yan N, Cheng H, Liu X, Chen F, Wang M. Lightweight privacy-preserving feature extraction for EEG signals under edge computing. IEEE Internet Things J. 2023;11(2):2520\u201333.","journal-title":"IEEE Internet Things J"},{"issue":"5","key":"470_CR16","doi-asserted-by":"publisher","first-page":"4258","DOI":"10.1109\/TDSC.2022.3218782","volume":"20","author":"M Wang","year":"2022","unstructured":"Wang M, Wang S, Hu J. Polycosgraph: a privacy-preserving cancelable EEG biometric system. IEEE Trans Dependable Secure Comput. 2022;20(5):4258\u201372.","journal-title":"IEEE Trans Dependable Secure Comput"},{"key":"470_CR17","doi-asserted-by":"crossref","unstructured":"Khanam T, Siuly S, Wang K, Wang H. Based-BCI technology. In: Health information science: 13th International Conference, HIS 2024, Hong Kong, December 8\u201310, 2024, Proceedings, vol. 15336. Springer Nature 2025. p. 209.","DOI":"10.1007\/978-981-96-5597-7_19"},{"key":"470_CR18","first-page":"84","volume-title":"International conference on web information systems engineering","author":"T Khanam","year":"2024","unstructured":"Khanam T, Siuly S, Wang K, Zheng Z. A privacy-preserving encryption framework for big data analysis. In: International conference on web information systems engineering. Springer; 2024. p. 84\u201394."},{"key":"470_CR19","first-page":"649","volume-title":"Annual cryptology conference","author":"M Abe","year":"2011","unstructured":"Abe M, Groth J, Haralambiev K, Ohkubo M. Optimal structure-preserving signatures in asymmetric bilinear groups. In: Annual cryptology conference. Springer; 2011. p. 649\u201366."},{"issue":"5","key":"470_CR20","doi-asserted-by":"publisher","first-page":"1829","DOI":"10.1109\/TCDS.2024.3386364","volume":"16","author":"MNA Tawhid","year":"2024","unstructured":"Tawhid MNA, Siuly S, Wang K, Wang H. Genet: a generic neural network for detecting various neurological disorders from EEG. IEEE Trans Cogn Dev Syst. 2024;16(5):1829\u201342.","journal-title":"IEEE Trans Cogn Dev Syst"},{"key":"470_CR21","doi-asserted-by":"publisher","first-page":"21954","DOI":"10.1109\/ACCESS.2017.2762418","volume":"5","author":"C Yin","year":"2017","unstructured":"Yin C, Zhu Y, Fei J, He X. A deep learning approach for intrusion detection using recurrent neural networks. IEEE Access. 2017;5:21954\u201361.","journal-title":"IEEE Access"},{"key":"470_CR22","doi-asserted-by":"crossref","unstructured":"Jia T, Meng L, Li S, Liu J, Wu D. Federated motor imagery classification for privacy-preserving brain-computer interfaces. IEEE Trans Neural Syst Rehabil Eng 2024.","DOI":"10.1109\/TNSRE.2024.3457504"},{"issue":"11","key":"470_CR23","doi-asserted-by":"publisher","first-page":"2115","DOI":"10.3390\/electronics14112115","volume":"14","author":"H-S Kim","year":"2025","unstructured":"Kim H-S, Kang J-W, Choi S-Y. Chatgpt vs. human journalists: analyzing news summaries through bertscore and moderation standards. Electronics. 2025;14(11):2115.","journal-title":"Electronics"},{"key":"470_CR24","doi-asserted-by":"crossref","unstructured":"Gu Z, Zhang L, Zhu X, Chen J, Huang W, Zhang Y, Wang S, Ye Z, Gao Y, Feng H, et al. Detectbench: Can large language model detect and piece together implicit evidence? arXiv preprint. 2024. arXiv:2406.12641.","DOI":"10.18653\/v1\/2024.findings-emnlp.11"},{"key":"470_CR25","first-page":"1","volume-title":"2024 International conference on current trends in advanced computing (ICCTAC)","author":"B Sindhu","year":"2024","unstructured":"Sindhu B, Prathamesh R, Sameera M, KumaraSwamy S. The evolution of large language model: models, applications and challenges. In: 2024 International conference on current trends in advanced computing (ICCTAC). IEEE; 2024. p. 1\u20138."},{"issue":"4","key":"470_CR26","doi-asserted-by":"publisher","first-page":"16","DOI":"10.1109\/MCI.2022.3199622","volume":"17","author":"S Li","year":"2022","unstructured":"Li S, Wu H, Ding L, Wu D. Meta-learning for fast and privacy-preserving source knowledge transfer of EEG-based BCIs. IEEE Comput Intell Mag. 2022;17(4):16\u201326.","journal-title":"IEEE Comput Intell Mag"},{"key":"470_CR27","first-page":"7470","volume-title":"ICASSP 2024\u20132024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","author":"Y-M Huang","year":"2024","unstructured":"Huang Y-M, Hung H-N, Tseng VS. Privacy-preserving attention-weighted multi-source domain adaptation for EEG motor imagery. In: ICASSP 2024\u20132024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE; 2024. p. 7470\u20134."}],"container-title":["Health Information Science and Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s13755-026-00470-x.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s13755-026-00470-x","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s13755-026-00470-x.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,8]],"date-time":"2026-07-08T09:43:09Z","timestamp":1783503789000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s13755-026-00470-x"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,8]]},"references-count":27,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,12]]}},"alternative-id":["470"],"URL":"https:\/\/doi.org\/10.1007\/s13755-026-00470-x","relation":{},"ISSN":["2047-2501"],"issn-type":[{"value":"2047-2501","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,8]]},"assertion":[{"value":"24 May 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"19 June 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"8 July 2026","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"The authors declare that they have no conflict of interest.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Conflict of interest"}},{"value":"This study used secondary data that is publicly available and freely accessible online. In the original data collection, all participants provided informed consent for their data to be used for research in a confidential manner. No participants in our study are directly connected to our research team, as this is a secondary analysis of de-identified data. All data were fully anonymized before we accessed them. Therefore, this research did not require additional ethical approval.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethical approval and consent to participate"}},{"value":"Not applicable.","order":4,"name":"Ethics","group":{"name":"EthicsHeading","label":"Consent for publication"}}],"article-number":"73"}}