Exploring EEG-based motor imagery decoding: a dual approach using spatial features and spectro-spatial Deep Learning model IFNet

• Published in: Frontiers in neuroinformatics Vol. 18; p. 1345425
• Main Authors: Juan, Javier V, Martínez, Rubén, Iáñez, Eduardo, Ortiz, Mario, Tornero, Jesús, Azorín, José M
• Format: Journal Article
• Language: English
• Published: Switzerland Frontiers Research Foundation 29.02.2024; Frontiers Media S.A
• Subjects: Algorithms; brain-machine interface (BMI); common spatial patterns filter bank (CSPFB); convolutional neural network (CNN); Cortex (motor); Cortex (temporal); Deep learning; deep learning (DL); Discriminant analysis; EEG; Electroencephalography; electroencephalography (EEG); Frequency; Mental task performance; Methods; motor imagery (MI); Motor task performance; Neural networks; Neurology; Neuroscience; Questionnaires; Recording equipment; Rehabilitation; Spatial discrimination learning; Temporal variations; Wavelet transforms; common spatial patterns filter bank (CSPFB); electroencephalography (EEG); brain-machine interface (BMI); convolutional neural network (CNN); IFNet; motor imagery (MI); linear discriminant analysis (LDA); deep learning (DL)
• ISSN: 1662-5196; 1662-5196
• DOI: 10.3389/fninf.2024.1345425

In recent years, the decoding of motor imagery (MI) from electroencephalography (EEG) signals has become a focus of research for brain-machine interfaces (BMIs) and neurorehabilitation. However, EEG signals present challenges due to their non-stationarity and the substantial presence of noise commonly found in recordings, making it difficult to design highly effective decoding algorithms. These algorithms are vital for controlling devices in neurorehabilitation tasks, as they activate the patient's motor cortex and contribute to their recovery. This study proposes a novel approach for decoding MI during pedalling tasks using EEG signals. A widespread approach is based on feature extraction using Common Spatial Patterns (CSP) followed by a linear discriminant analysis (LDA) as a classifier. The first approach covered in this work aims to investigate the efficacy of a task-discriminative feature extraction method based on CSP filter and LDA classifier. Additionally, the second alternative hypothesis explores the potential of a spectro-spatial Convolutional Neural Network (CNN) to further enhance the performance of the first approach. The proposed CNN architecture combines a preprocessing pipeline based on filter banks in the frequency domain with a convolutional neural network for spectro-temporal and spectro-spatial feature extraction. To evaluate the approaches and their advantages and disadvantages, EEG data has been recorded from several able-bodied users while pedalling in a cycle ergometer in order to train motor imagery decoding models. The results show levels of accuracy up to 80% in some cases. The CNN approach shows greater accuracy despite higher instability.