A multiscale siamese convolutional neural network with cross-channel fusion for motor imagery decoding

Functional connectivity learning via Siamese-based SPD matrix representation of brain imaging data

Measurement of brain functional connectivity has become a dominant approach to explore the interaction dynamics between brain regions of subjects under examination. Conventional functional connectivity measures largely originate from deterministic models on empirical analysis, usually demanding application-specific settings (e.g., Pearson's Correlation and Mutual Information). To bridge the technical gap, this study proposes a Siamese-based Symmetric Positive Definite (SPD) Matrix Representation framework (SiameseSPD-MR) to derive the functional connectivity of brain imaging data (BID) such as Electroencephalography (EEG), thus the alternative application-independent measure (in the form of SPD matrix) can be automatically learnt: (1) SiameseSPD-MR first exploits graph convolution to extract the representative features of BID with the adjacency matrix computed considering the anatomical structure; (2) Adaptive Gaussian kernel function then applies to obtain the functional connectivity representations from the deep features followed by SPD matrix transformation to address the intrinsic functional characteristics; and (3) Two-branch (Siamese) networks are combined via an element-wise product followed by a dense layer to derive the similarity between the pairwise inputs. Experimental results on two EEG datasets (autism spectrum disorder, emotion) indicate that (1) SiameseSPD-MR can capture more significant differences in functional connectivity between neural states than the state-of-the-art counterparts do, and these findings properly highlight the typical EEG characteristics of ASD subjects, and (2) the obtained functional connectivity representations conforming to the proposed measure can act as meaningful markers for brain network analysis and ASD discrimination.

Filter bank sinc-convolutional network with channel self-attention for high performance motor imagery decoding

Objective.Motor Imagery Brain-Computer Interface (MI-BCI) is an active Brain-Computer Interface (BCI) paradigm focusing on the identification of motor intention, which is one of the most important non-invasive BCI paradigms. In MI-BCI studies, deep learning-based methods (especially lightweight networks) have attracted more attention in recent years, but the decoding performance still needs further improving.Approach.To solve this problem, we designed a filter bank structure with sinc-convolutional layers for spatio-temporal feature extraction of MI-electroencephalography in four motor rhythms. The Channel Self-Attention method was introduced for feature selection based on both global and local information, so as to build a model called Filter Bank Sinc-convolutional Network with Channel Self-Attention for high performance MI-decoding. Also, we proposed a data augmentation method based on multivariate empirical mode decomposition to improve the generalization capability of the model.Main results.We performed an intra-subject evaluation experiment on unseen data of three open MI datasets. The proposed method achieved mean accuracy of 78.20% (4-class scenario) on BCI Competition IV IIa, 87.34% (2-class scenario) on BCI Competition IV IIb, and 72.03% (2-class scenario) on Open Brain Machine Interface (OpenBMI) dataset, which are significantly higher than those of compared deep learning-based methods by at least 3.05% (p= 0.0469), 3.18% (p= 0.0371), and 2.27% (p= 0.0024) respectively.Significance.This work provides a new option for deep learning-based MI decoding, which can be employed for building BCI systems for motor rehabilitation.

An Intelligent Motor Imagery Detection System Using Electroencephalography with Adaptive Wavelets

Classification of motor imagery (MI) tasks provides a robust solution for specially-abled people to connect with the milieu for brain-computer interface. Precise selection of uniform tuning parameters of tunable Q wavelet transform (TQWT) for electroencephalography (EEG) signals is arduous. Therefore, this paper proposes robust TQWT for automatically selecting optimum tuning parameters to decompose non-stationary EEG signals accurately. Three evolutionary optimization algorithms are explored for automating the tuning parameters of robust TQWT. The fitness function of the mean square error of decomposition is used. This paper also exploits channel selection using a Laplacian score for dominant channel selection. Important features elicited from sub-bands of robust TQWT are classified using different kernels of the least square support vector machine classifier. The radial basis function kernel has provided the highest accuracy of 99.78%, proving that the proposed method is superior to other state-of-the-art using the same database.

Considerate motion imagination classification method using deep learning

In order to improve the classification accuracy of motion imagination, a considerate motion imagination classification method using deep learning is proposed. Specifically, based on a graph structure suitable for electroencephalography as input, the proposed model can accurately represent the distribution of electroencephalography electrodes in non-Euclidean space and fully consider the spatial correlation between electrodes. In addition, the spatial-spectral-temporal multi-dimensional feature information was extracted from the spatial-temporal graph representation and spatial-spectral graph representation transformed from the original electroencephalography signal using the dual branch architecture. Finally, the attention mechanism and global feature aggregation module were designed and combined with graph convolution to adaptively capture the dynamic correlation intensity and effective feature of electroencephalography signals in various dimensions. A series of contrast experiments and ablation experiments on several different public brain-computer interface datasets demonstrated that the excellence of proposed method. It is worth mentioning that, the proposed model is a general framework for the classification of electroencephalography signals, which is suitable for emotion recognition, sleep staging and other fields based on electroencephalography research. Moreover, the model has the potential to be applied in the medical field of motion imagination rehabilitation in real life.

Multi-band network fusion for Alzheimer's disease identification with functional MRI

INTRODUCTION

The analysis of functional brain networks (FBNs) has become a promising and powerful tool for auxiliary diagnosis of brain diseases, such as Alzheimer's disease (AD) and its prodromal stage. Previous studies usually estimate FBNs using full band Blood Oxygen Level Dependent (BOLD) signal. However, a single band is not sufficient to capture the diagnostic and prognostic information contained in multiple frequency bands.

METHOD

To address this issue, we propose a novel multi-band network fusion framework (MBNF) to combine the various information (e.g., the diversification of structural features) of multi-band FBNs. We first decompose the BOLD signal adaptively into two frequency bands named high-frequency band and low-frequency band by the ensemble empirical mode decomposition (EEMD). Then the similarity network fusion (SNF) is performed to blend two networks constructed by two frequency bands together into a multi-band fusion network. In addition, we extract the features of the fused network towards a better classification performance.

RESULT

To verify the validity of the scheme, we conduct our MBNF method on the public ADNI database for identifying subjects with AD/MCI from normal controls.

DISCUSSION

Experimental results demonstrate that the proposed scheme extracts rich multi-band network features and biomarker information, and also achieves better classification accuracy.