Noise-Factorized Disentangled Representation Learning for Generalizable Motor Imagery EEG Classification

• Published in: IEEE journal of biomedical and health informatics Vol. 28; no. 2; pp. 765 - 776
• Main Authors: Han, Jinpei, Gu, Xiao, Yang, Guang-Zhong, Lo, Benny
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.02.2024; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Benchmarking; Brain; Brain modeling; Brain-Computer Interfaces; Classification; Computer applications; Disentangled representation learning; EEG; Electroencephalography; Electroencephalography - methods; Feature extraction; generalization; Human-computer interface; Humans; Image classification; Image reconstruction; Imagination; Implants; Machine learning; Mental task performance; motor imagery EEG; Motor skill learning; Representation learning; Representations; Task analysis; Training
• ISSN: 2168-2194; 2168-2208; 2168-2208
• DOI: 10.1109/JBHI.2023.3337072

Motor Imagery (MI) Electroencephalography (EEG) is one of the most common Brain-Computer Interface (BCI) paradigms that has been widely used in neural rehabilitation and gaming. Although considerable research efforts have been dedicated to developing MI EEG classification algorithms, they are mostly limited in handling scenarios where the training and testing data are not from the same subject or session. Such poor generalization capability significantly limits the realization of BCI in real-world applications. In this paper, we proposed a novel framework to disentangle the representation of raw EEG data into three components, subject/session-specific, MI-task-specific, and random noises, so that the subject/session-specific feature extends the generalization capability of the system. This is realized by a joint discriminative and generative framework, supported by a series of fundamental training losses and training strategies. We evaluated our framework on three public MI EEG datasets, and detailed experimental results show that our method can achieve superior performance by a large margin compared to current state-of-the-art benchmark algorithms.