Msst-eegnet: multi-scale spatio-temporal feature extraction using inception and temporal pyramid pooling for motor imagery classification

• Published in: Cognitive neurodynamics Vol. 19; no. 1; p. 150
• Main Authors: Mishra, Rashmi, Agrawal, R. K., Kirar, Jyoti Singh
• Format: Journal Article
• Language: English
• Published: Dordrecht Springer Netherlands 01.12.2025; Springer Nature B.V
• Subjects: Accuracy; Artificial Intelligence; Biochemistry; Biochips; Biomedical and Life Sciences; Biomedicine; Brain; Brain research; Classification; Cognitive Psychology; Competition; Computer applications; Computer Science; Convolution; Datasets; Deep learning; Electroencephalography; Feature extraction; Fourier transforms; Human-computer interface; Imagery; Implants; Machine learning; Mental task performance; Modules; Motor task performance; Neurosciences; Research Article; Signal to noise ratio; Statistical analysis; Statistical tests; Temporal variations; Transfer learning; Inception module; Dilated convolution; Brain computer interface; Multi-class motor imagery classification; Temporal pyramid pooling
• ISSN: 1871-4080; 1871-4099
• DOI: 10.1007/s11571-025-10337-8

Motor imagery classification is an essential component of Brain-computer interface systems to interpret and recognize brain signals generated during the visualization of motor imagery tasks by a subject. The objective of this work is to develop a novel DL model to extract discriminative features for better generalization performance to recognize motor imagery tasks. This paper presents a novel Multi-scale spatio-temporal network (MSST-EEGNet) to extract discriminative temporal, spectral, and spatial features for motor imagery task classification. The proposed MSST-EEGNet model includes three modules namely the inception module with dilated convolution, the temporal pyramid pooling module, and the classification module. Multi-scale temporal features along with spatial features are extracted using the inception block with the dilated convolution module. A set of multi-level fine-grained and coarse-grained features are extracted using a temporal pyramid pooling module. Further, categorical cross-entropy in combination with center loss is used as a loss function. Experiments are carried out on three benchmark datasets including the BCI Competition IV-2a dataset, the BCI Competition IV-2b dataset, and the OpenBMI dataset. The evaluation results shows that the proposed MSST-EEGNet model outperforms eight existing DL models in terms of classification accuracy for subject-specific and cross-session settings. It also outperforms eight existing DL models and six existing transfer-learning models for cross-subject setting. For the subject-specific classification the proposed MSST-EEGNet model achieved an accuracy of 0.8426 ± 0.1061, 0.7779 ± 0.0938, and 0.7365 ± 0.1477 on the BCI Competition IV-2a dataset, the BCI Competition IV-2b dataset, and the OpenBMI dataset respectively. For the cross-session setting, the proposed MSST-EEGNet model achieved an accuracy of 0.7709 ± 0.1098, 0.7524 ± 0.1017, and 0.6860 ± 0.0990 on the BCI Competition IV-2a dataset, the BCI Competition IV-2b dataset, and the OpenBMI dataset respectively. For the cross-subject setting, the proposed MSST-EEGNet model achieved an accuracy of 0.7288 ± 0.0730, 0.8161 ± 0.963, and 0.7075 ± 0.0746 on the BCI Competition IV-2a dataset, the BCI Competition IV-2b dataset, and the OpenBMI dataset respectively. Furthermore, a non-parametric Friedman statistical test demonstrates statistically significant superior performance of the proposed MSST-EEGNet model over the existing models.