Hierarchical Deep Spatial-Spectral Architecture for Motor Imagery Eeg Processing

View at publisher Summarise with VIRAT
Authors: Nirmala Priyadharshini P, N SathishKumar, S. Shankar, D Vidhya, Nithya V, S.Shamna Parveen
Cited by: 0
References: 16

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Motor image based electroencephalography is an important technique used in brain computer interface systems to interpret brain activity related to imagined movements. yet accurate classification of MI EEG signals is difficult due to the complex spatial and frequency variations across subjects and channels. This paper presents a spatial-spectral that multi-stage to the neural architecture designed to effectively extract both spatial and frequency features for reliable MI EEG interpretation. The proposed model combines frequency band decomposition with spatial filtering in multiple stages to capture detailed signal patterns. Prior to final classification, the retrieved spatial-spectral characteristics are integrated using a hierarchical fusion layer. In terms of classification accuracy, tests on popular motor imaging datasets demonstrate that the proposed model performs better than traditional and one-stage deep learning methods. The results show that integrating to the spectral and spatial representations enhances participant consistency and robustness, contributing to the effectiveness of EEG-based motor imagery decoding.

Fields of Study & Subjects

EEG and Brain-Computer Interfaces Muscle activation and electromyography studies Emotion and Mood Recognition

Publication details

Journal / source
International Conference on Communication, Computing and Electronics Systems (ICCCES)
Pages
632–637
Publication date
21 January 2026
Language
EN