Huang, XiuyuChoi, Kup-SzeZhou, NanZhang, YuanpengChen, BadongPedrycz, Witold2024-05-192024-05-1920242379-89202379-8939https://doi.org10.1109/TCDS.2023.3279262https://hdl.handle.net/20.500.12713/5115Electroencephalography (EEG) data across multiple individuals have a high variance. Directly using the data to train a deep learning (DL) model usually degrades the performance. To address this issue, we propose a shallow Inception domain adaptation framework to extract informative deep features from data of multiple subjects for accurate motor imagery (MI) recognition. To our best knowledge, the Inception architecture in DL is combined with a domain adaptation (DA) scheme for the first time for the MI classification task. The approach contains two compact Inception blocks that decode temporal features in different scales. In addition, we jointly optimize a novel combined loss function to reduce both marginal and class conditional discrepancies caused by the multimodal structure of EEG signals. The DA-based loss enables Inception blocks to take full advantage of their learning abilities to capture discriminative patterns of MI data from multiple subjects instead of relying on the target user only. To demonstrate the effectiveness of our approach, we conduct substantial experiments on two well-known data sets, brain-computer interface competition IV-2a and competition IV-2b. Results show that our model achieves better performance than state-of-the-art strategies. The proposed model is able to extract informative features from high-variant EEG data collected from different individuals and achieves accurate MI classifications.eninfo:eu-repo/semantics/closedAccessFeature ExtractionElectroencephalographyBrain ModelingTrainingTask AnalysisData ModelsDeep LearningDeep Neural NetworkDomain Adaptation (Da)InceptionMotor Imagery (Mi)Article162521533WOS:0011978610000172-s2.0-85161050222N/A10.1109/TCDS.2023.3279262Q1