Federated Transfer Learning for EEG Signal Classification
WeBank Co., Ltd.(微众银行);Hong Kong University of Science and Technology(香港科技大学);Southern University of Science and Technology(南方科技大学);Nanyang Technological University(南洋理工大学)
Ce Ju, Dashan Gao, Ravikiran Mane, Ben Tan, Yang Liu, Cuntai Guan
The success of deep learning (DL) methods in the Brain-Computer Interfaces (BCI) field for classification of electroencephalographic (EEG) recordings has been restricted by the lack of large datasets. Privacy concerns associated with EEG signals limit the possibility of constructing a large EEG-BCI dataset by the conglomeration of multiple small ones for jointly training machine learning models. Hence, in this paper, we propose a novel privacy-preserving DL architecture named federated transfer learning (FTL) for EEG classification that is based on the federated learning framework. Working with the single-trial covariance matrix, the proposed architecture extracts common discriminative information from multi-subject EEG data with the help of domain adaptation techniques. We evaluate the performance of the proposed architecture on the PhysioNet dataset for 2-class motor imagery classification. While avoiding the actual data sharing, our FTL approach achieves 2% higher classification accuracy in a subject-adaptive analysis. Also, in the absence of multi-subject data, our architecture provides 6% better accuracy compared to other state-of-the-art DL architectures.
Federated Few-Shot Learning for Epileptic Seizure Detection Under Privacy Constraints
Pro2Future GmbH(Pro2Future公司);Institute for Machine Learning(机器学习研究所);Johannes Kepler University(约瑟夫·约瑟夫·凯普勒大学);Institute of Pervasive Computing(普及计算研究所)
Interpretable Multimodal EEG-ECG Framework for Neonatal Seizure Prediction and Patient-Adaptive Transfer Learning
Department of Electronic and Telecommunication Engineering, University of Moratuwa, Sri Lanka(电子与电信工程系,乌尔塔瓦大学,斯里兰卡);Department of Biomedical Engineering, University of Melbourne, Australia(生物医学工程系,墨尔本大学,澳大利亚);Department of Electrical and Computer Engineering, University of British Columbia, Canada(电气与计算机工程系,不列颠哥伦比亚大学,加拿大);Department of Pediatrics, Faculty of Medicine, University of Colombo, Sri Lanka(儿科系,医学院,科伦坡大学,斯里兰卡)
BrainGPT: Unleashing the Potential of EEG Generalist Foundation Model by Autoregressive Pre-training
Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所);School of Artificial Intelligence, University of Chinese Academy of Sciences(中国科学院大学人工智能学院);Beijing Jiaotong University(北京交通大学)
FetalSleepNet: A Transfer Learning Framework with Spectral Equalisation Domain Adaptation for Fetal Sleep Stage Classification
Department of Biomedical Informatics, Emory University(埃默里大学生物医学信息学系);Ritchie Centre, Hudson Institute of Medical Research(哈德逊医学研究所里奇中心);Department of Biomedical Engineering, Georgia Institute of Technology(佐治亚理工学院生物医学工程系);Department of Obstetrics and Gynaecology, Monash University(莫纳什大学妇产科学系);Department of Electrical and Computer Systems Engineering, Monash University(莫纳什大学电气与计算机系统工程系)
A Novel TSK Fuzzy System Incorporating Multi-view Collaborative Transfer Learning for Personalized Epileptic EEG Detection
School of Artificial Intelligence and Computer Science, Jiangnan University and Engineering Research Center of Intelligent Technology for Healthcare, Ministry of Education(江南大学人工智能与计算机科学学院,智能医疗技术工程研究中心,教育部);School of Computer and Information Engineering, Institute for Artificial Intelligence, Shanghai Polytechnic University(上海理工大学计算机与信息工程学院,人工智能研究院);School of Internet of Things Engineering, Wuxi University(无锡大学物联网工程学院)
Inter-subject Deep Transfer Learning for Motor Imagery EEG Decoding
Imperial College London(帝国理工学院)
卷积神经网络(CNN)已经成为解码EEG的强大技术,并且已经成为运动想象EEG脑机接口(BCI)解码的基准。然而,在不降低个体性能的情况下对多个受试者的EEG进行训练CNN仍然具有挑战性。这被称为负迁移问题,即从不同的分布中学习会导致CNN错误地表示它们中的每一个,而不是学习更丰富的表示。因此,CNN不能直接使用多个受试者的EEG来直接增强模型性能。为了解决这个问题,我们将深度迁移学习技术扩展到EEG多主题训练案例。我们提出了一种多分支深度传输网络,即分离-公共-分离网络(SCSN),该网络基于为单个主题拆分网络的特征提取器。我们还探讨了将最大均值差异(MMD)应用于SCSN(SCSN-MMD)的可能性,以更好地对齐来自各个特征提取器的特征分布。在BCI Competition IV 2a数据集(BCICIV 2a数据集)和我们的在线记录数据集上对所提出的网络进行了评估。结果表明,提出的SCSN(81.8%,53.2%)和SCSN-MMD(81.8%,54.8%)优于基准CNN(73.4%,48.8%)在两个数据集上使用多个主题。我们提出的网络显示了利用更大的多学科数据集训练EEG解码器而不受负传递影响的潜力。
EEGReXferNet: A Lightweight Gen-AI Framework for EEG Subspace Reconstruction via Cross-Subject Transfer Learning and Channel-Aware Embedding
IUCRC BRAIN Center, Cullen College of Engineering, University of Houston(IUCRC BRAIN中心、工程学院、休斯顿大学);Department of Computer Engineering, Purdue University(计算机工程系、普渡大学)
Revisiting Euclidean Alignment for Transfer Learning in EEG-Based Brain-Computer Interfaces
Ministry of Education Key Laboratory of Image Processing and Intelligent Control, Huazhong University of Science and Technology, Wuhan, 430074 China(教育部长图像处理与智能控制重点实验室,华中科技大学,武汉,430074中国);Zhongguancun Academy, Beijing, 100084 China(中关村学院,北京,100084中国)
2021 BEETL Competition: Advancing Transfer Learning for Subject Independence & Heterogenous EEG Data Sets
Imperial College London(帝国理工学院);University of Bayreuth(拜罗伊特大学);University of Vienna(维也纳大学);Universite Paris-Saclay(巴黎萨克雷大学);Inria(法国国家信息与自动化研究所);CEA(法国原子能和替代能源委员会);UVSQ(凡尔赛大学);Reality Labs;University of Bordeaux(波尔多大学);Cogitat Ltd.(Cogitat有限公司);Aristotle University of Thessaloniki(塞萨洛尼基亚里士多德大学);National and Kapodistrian University of Athens(雅典国立卡波季斯特里安大学);DCS Corporation(DCS公司);DEVCOM Army Research Laboratory(DEVCOM陆军研究实验室);Univ. Grenoble Alpes(格勒诺布尔阿尔卑斯大学);Polish National Institute for Machine Learning (OPIUM)(波兰国家机器学习研究所(OPIUM))
M3D: Manifold-based Domain Adaptation with Dynamic Distribution for Non-Deep Transfer Learning in Cross-subject and Cross-session EEG-based Emotion Recognition
School of Biomedical Engineering, Medical School, Shenzhen University(生物医学工程学院,医学学院,深圳大学);School of Public Health, Shenzhen University Medical School(公共卫生学院,深圳大学医学学院);Department of Radiology, Nanshan Hospital of Shenzhen University(放射科,深圳大学南山医院);School of Mathematical Sciences, Shenzhen University(数学科学学院,深圳大学);Shien-Ming Wu School of Intelligent Engineering, South China University of Technology(申明武智能工程学院,华南理工大学);Institute for Super Robotics, Guangzhou, Guangdong, China(超机器人研究所,广州,广东,中国)
MS-MDA: Multisource Marginal Distribution Adaptation for Cross-subject and Cross-session EEG Emotion Recognition
HwaMei Hospital, University of Chinese Academy of Sciences(中国科学院大学华美医院);Ningbo Institute of Life and Health Industry, University of Chinese Academy of Sciences(中国科学院大学宁波生命与健康产业研究院);Anhui University(安徽大学);Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所)
Differential Private Federated Transfer Learning for Mental Health Monitoring in Everyday Settings: A Case Study on Stress Detection
University of California, Irvine(加利福尼亚大学欧文分校);School of Nursing, University of California, Irvine(加利福尼亚大学欧文分校护理学院);Institute for Future Health, University of California, Irvine(加利福尼亚大学欧文分校未来健康研究所)