Mamba with split-based pyramidal convolution and Kolmogorov-Arnold network-channel-spatial attention for electroencephalogram classification
Deep learning is widely used in brain electrical signal studies, among which the brain–computer interface is an important direction. Deep learning can effectively improve the performance of BCI machines, which is of great medical and commercial value. This paper introduces an efficient deep learning...
Saved in:
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2025-04-01
|
| Series: | Frontiers in Sensors |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fsens.2025.1548729/full |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849701593205702656 |
|---|---|
| author | Zhe Li Zhe Li |
| author_facet | Zhe Li Zhe Li |
| author_sort | Zhe Li |
| collection | DOAJ |
| description | Deep learning is widely used in brain electrical signal studies, among which the brain–computer interface is an important direction. Deep learning can effectively improve the performance of BCI machines, which is of great medical and commercial value. This paper introduces an efficient deep learning model for classifying brain electrical signals based on a Mamba structure enhanced with split-based pyramidal convolution (PySPConv) and Kolmogorov-Arnold network (KAN)-channel-spatial attention (KSA) mechanisms. Incorporating KANs into the attention module of the proposed KSA-Mamba-PySPConv model better approximates the sample function while obtaining local network features. PySPConv, on the other hand, swiftly and efficiently extracts multi-scale fusion features from input data. This integration allows the model to reinforce feature extraction at each layer in Mamba’s structure. The model achieves a 96.76% accuracy on the eegmmidb dataset and demonstrates state-of-the-art performance across metrics such as the F1 score, precision, and recall. KSA-Mamba-PySPConv promises to be an effective tool in electroencephalogram classification in brain–computer interface systems. |
| format | Article |
| id | doaj-art-4c4c327ea48c482da32c09d9f4f2e382 |
| institution | DOAJ |
| issn | 2673-5067 |
| language | English |
| publishDate | 2025-04-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Sensors |
| spelling | doaj-art-4c4c327ea48c482da32c09d9f4f2e3822025-08-20T03:17:54ZengFrontiers Media S.A.Frontiers in Sensors2673-50672025-04-01610.3389/fsens.2025.15487291548729Mamba with split-based pyramidal convolution and Kolmogorov-Arnold network-channel-spatial attention for electroencephalogram classificationZhe Li0Zhe Li1Hankou University, Jiangxiazhen, ChinaWuhan University of Engineering Science, Wuhan, ChinaDeep learning is widely used in brain electrical signal studies, among which the brain–computer interface is an important direction. Deep learning can effectively improve the performance of BCI machines, which is of great medical and commercial value. This paper introduces an efficient deep learning model for classifying brain electrical signals based on a Mamba structure enhanced with split-based pyramidal convolution (PySPConv) and Kolmogorov-Arnold network (KAN)-channel-spatial attention (KSA) mechanisms. Incorporating KANs into the attention module of the proposed KSA-Mamba-PySPConv model better approximates the sample function while obtaining local network features. PySPConv, on the other hand, swiftly and efficiently extracts multi-scale fusion features from input data. This integration allows the model to reinforce feature extraction at each layer in Mamba’s structure. The model achieves a 96.76% accuracy on the eegmmidb dataset and demonstrates state-of-the-art performance across metrics such as the F1 score, precision, and recall. KSA-Mamba-PySPConv promises to be an effective tool in electroencephalogram classification in brain–computer interface systems.https://www.frontiersin.org/articles/10.3389/fsens.2025.1548729/fullmambaKolmogorov-Arnold networkelectroencephalogramdeep learningBCI |
| spellingShingle | Zhe Li Zhe Li Mamba with split-based pyramidal convolution and Kolmogorov-Arnold network-channel-spatial attention for electroencephalogram classification Frontiers in Sensors mamba Kolmogorov-Arnold network electroencephalogram deep learning BCI |
| title | Mamba with split-based pyramidal convolution and Kolmogorov-Arnold network-channel-spatial attention for electroencephalogram classification |
| title_full | Mamba with split-based pyramidal convolution and Kolmogorov-Arnold network-channel-spatial attention for electroencephalogram classification |
| title_fullStr | Mamba with split-based pyramidal convolution and Kolmogorov-Arnold network-channel-spatial attention for electroencephalogram classification |
| title_full_unstemmed | Mamba with split-based pyramidal convolution and Kolmogorov-Arnold network-channel-spatial attention for electroencephalogram classification |
| title_short | Mamba with split-based pyramidal convolution and Kolmogorov-Arnold network-channel-spatial attention for electroencephalogram classification |
| title_sort | mamba with split based pyramidal convolution and kolmogorov arnold network channel spatial attention for electroencephalogram classification |
| topic | mamba Kolmogorov-Arnold network electroencephalogram deep learning BCI |
| url | https://www.frontiersin.org/articles/10.3389/fsens.2025.1548729/full |
| work_keys_str_mv | AT zheli mambawithsplitbasedpyramidalconvolutionandkolmogorovarnoldnetworkchannelspatialattentionforelectroencephalogramclassification AT zheli mambawithsplitbasedpyramidalconvolutionandkolmogorovarnoldnetworkchannelspatialattentionforelectroencephalogramclassification |