While significant advancements in artificial intelligence (AI) have catalyzed progress across various domains, its full potential in understanding visual perception remains underexplored. We propose an artificial neural network dubbed VISION, an acronym for "Visual Interface System for Imaging Output of Neural activity," to mimic the human brain and show how it can foster neuroscientific inquiries. Using visual and contextual inputs, this multimodal model predicts the brain's functional magnetic resonance imaging (fMRI) scan response to natural images. VISION successfully predicts human hemodynamic responses as fMRI voxel values to visual inputs with an accuracy exceeding state-of-the-art performance by 45%. We further probe the trained networks to reveal representational biases in different visual areas, generate experimentally testable hypotheses, and formulate an interpretable metric to associate these hypotheses with cortical functions. With both a model and evaluation metric, the cost and time burdens associated with designing and implementing functional analysis on the visual cortex could be reduced. Our work suggests that the evolution of computational models may shed light on our fundamental understanding of the visual cortex and provide a viable approach toward reliable brain-machine interfaces.
This paper presents a source free domain adaptation method for steady-state visually evoked potentials (SSVEP) based brain-computer interface (BCI) spellers. SSVEP-based BCI spellers assist individuals experiencing speech difficulties by enabling them to communicate at a fast rate. However, achieving a high information transfer rate (ITR) in most prominent methods requires an extensive calibration period before using the system, leading to discomfort for new users. We address this issue by proposing a novel method that adapts a powerful deep neural network (DNN) pre-trained on data from source domains (data from former users or participants of previous experiments) to the new user (target domain), based only on the unlabeled target data. This adaptation is achieved by minimizing our proposed custom loss function composed of self-adaptation and local-regularity terms. The self-adaptation term uses the pseudo-label strategy, while the novel local-regularity term exploits the data structure and forces the DNN to assign similar labels to adjacent instances. The proposed method priorities user comfort by removing the burden of calibration while maintaining an excellent character identification accuracy and ITR. In particular, our method achieves striking 201.15 bits/min and 145.02 bits/min ITRs on the benchmark and BETA datasets, respectively, and outperforms the state-of-the-art alternatives. Our code is available at https://github.com/osmanberke/SFDA-SSVEP-BCI
EEG-based recognition of activities and states involves the use of prior neuroscience knowledge to generate quantitative EEG features, which may limit BCI performance. Although neural network-based methods can effectively extract features, they often encounter issues such as poor generalization across datasets, high predicting volatility, and low model interpretability. Hence, we propose a novel lightweight multi-dimensional attention network, called LMDA-Net. By incorporating two novel attention modules designed specifically for EEG signals, the channel attention module and the depth attention module, LMDA-Net can effectively integrate features from multiple dimensions, resulting in improved classification performance across various BCI tasks. LMDA-Net was evaluated on four high-impact public datasets, including motor imagery (MI) and P300-Speller paradigms, and was compared with other representative models. The experimental results demonstrate that LMDA-Net outperforms other representative methods in terms of classification accuracy and predicting volatility, achieving the highest accuracy in all datasets within 300 training epochs. Ablation experiments further confirm the effectiveness of the channel attention module and the depth attention module. To facilitate an in-depth understanding of the features extracted by LMDA-Net, we propose class-specific neural network feature interpretability algorithms that are suitable for event-related potentials (ERPs) and event-related desynchronization/synchronization (ERD/ERS). By mapping the output of the specific layer of LMDA-Net to the time or spatial domain through class activation maps, the resulting feature visualizations can provide interpretable analysis and establish connections with EEG time-spatial analysis in neuroscience. In summary, LMDA-Net shows great potential as a general online decoding model for various EEG tasks.
Electroencephalography (EEG) plays a significant role in the Brain Computer Interface (BCI) domain, due to its non-invasive nature, low cost, and ease of use, making it a highly desirable option for widespread adoption by the general public. This technology is commonly used in conjunction with deep learning techniques, the success of which is largely dependent on the quality and quantity of data used for training. To address the challenge of obtaining sufficient EEG data from individual participants while minimizing user effort and maintaining accuracy, this study proposes an advanced methodology for data augmentation: generating synthetic EEG data using denoising diffusion probabilistic models. The synthetic data are generated from electrode-frequency distribution maps (EFDMs) of emotionally labeled EEG recordings. To assess the validity of the synthetic data generated, both a qualitative and a quantitative comparison with real EEG data were successfully conducted. This study opens up the possibility for an open\textendash source accessible and versatile toolbox that can process and generate data in both time and frequency dimensions, regardless of the number of channels involved. Finally, the proposed methodology has potential implications for the broader field of neuroscience research by enabling the creation of large, publicly available synthetic EEG datasets without privacy concerns.
When dealing with electro or magnetoencephalography records, many supervised prediction tasks are solved by working with covariance matrices to summarize the signals. Learning with these matrices requires using Riemanian geometry to account for their structure. In this paper, we propose a new method to deal with distributions of covariance matrices and demonstrate its computational efficiency on M/EEG multivariate time series. More specifically, we define a Sliced-Wasserstein distance between measures of symmetric positive definite matrices that comes with strong theoretical guarantees. Then, we take advantage of its properties and kernel methods to apply this distance to brain-age prediction from MEG data and compare it to state-of-the-art algorithms based on Riemannian geometry. Finally, we show that it is an efficient surrogate to the Wasserstein distance in domain adaptation for Brain Computer Interface applications.
In neural decoding research, one of the most intriguing topics is the reconstruction of perceived natural images based on fMRI signals. Previous studies have succeeded in re-creating different aspects of the visuals, such as low-level properties (shape, texture, layout) or high-level features (category of objects, descriptive semantics of scenes) but have typically failed to reconstruct these properties together for complex scene images. Generative AI has recently made a leap forward with latent diffusion models capable of generating high-complexity images. Here, we investigate how to take advantage of this innovative technology for brain decoding. We present a two-stage scene reconstruction framework called ``Brain-Diffuser''. In the first stage, starting from fMRI signals, we reconstruct images that capture low-level properties and overall layout using a VDVAE (Very Deep Variational Autoencoder) model. In the second stage, we use the image-to-image framework of a latent diffusion model (Versatile Diffusion) conditioned on predicted multimodal (text and visual) features, to generate final reconstructed images. On the publicly available Natural Scenes Dataset benchmark, our method outperforms previous models both qualitatively and quantitatively. When applied to synthetic fMRI patterns generated from individual ROI (region-of-interest) masks, our trained model creates compelling ``ROI-optimal'' scenes consistent with neuroscientific knowledge. Thus, the proposed methodology can have an impact on both applied (e.g. brain-computer interface) and fundamental neuroscience.
Most of the Brain-Computer Interface (BCI) publications, which propose artificial neural networks for Motor Imagery (MI) Electroencephalography (EEG) signal classification, are presented using one of the BCI Competition datasets. However, these databases contain MI EEG data from less than or equal to 10 subjects . In addition, these algorithms usually include only bandpass filtering to reduce noise and increase signal quality. In this article, we compared 5 well-known neural networks (Shallow ConvNet, Deep ConvNet, EEGNet, EEGNet Fusion, MI-EEGNet) using open-access databases with many subjects next to the BCI Competition 4 2a dataset to acquire statistically significant results. We removed artifacts from the EEG using the FASTER algorithm as a signal processing step. Moreover, we investigated whether transfer learning can further improve the classification results on artifact filtered data. We aimed to rank the neural networks; therefore, next to the classification accuracy, we introduced two additional metrics: the accuracy improvement from chance level and the effect of transfer learning. The former can be used with different class-numbered databases, while the latter can highlight neural networks with sufficient generalization abilities. Our metrics showed that the researchers should not avoid Shallow ConvNet and Deep ConvNet because they can perform better than the later published ones from the EEGNet family.
Artificial intelligence has made significant advances in recent years and this has had an impact on the field of neuroscience. As a result, different architectures have been implemented to extract features from EEG signals in real time. However, the use of such architectures requires a lot of computing power. As a result, EEG devices typically act only as transmitters of EEG data, with the actual data processing taking place in a third-party device. That's expensive and not compact. In this paper, we present a shield that allows a single-board computer, the Jetson Nano from Nvidia, to be converted into a brain-computer interface and, most importantly, the Jetson Nano's capabilities allow machine learning tools to be used directly on the data collection device. Here we present the test results of the developed device. https://github.com/HackerBCI/EEG-with-JetsonNano
Steady-state visual evoked potentials (SSVEPs) based brain-computer interface (BCI) has received considerable attention due to its high information transfer rate (ITR) and available quantity of targets. However, the performance of frequency identification methods heavily hinges on the amount of user calibration data and data length, which hinders the deployment in real-world applications. Recently, generative adversarial networks (GANs)-based data generation methods have been widely adopted to create synthetic electroencephalography (EEG) data, holds promise to address these issues. In this paper, we proposed a GAN-based end-to-end signal transformation network for Time-window length Extension, termed as TEGAN. TEGAN transforms short-length SSVEP signals into long-length artificial SSVEP signals. By incorporating a novel U-Net generator architecture and an auxiliary classifier into the network architecture, the TEGAN could produce conditioned features in the synthetic data. Additionally, we introduced a two-stage training strategy and the LeCam-divergence regularization term to regularize the training process of GAN during the network implementation. The proposed TEGAN was evaluated on two public SSVEP datasets (a 4-class dataset and a 12-class dataset). With the assistance of TEGAN, the performance of traditional frequency recognition methods and deep learning-based methods have been significantly improved under limited calibration data. And the classification performance gap of various frequency recognition methods has been narrowed. This study substantiates the feasibility of the proposed method to extend the data length for short-time SSVEP signals for developing a high-performance BCI system. The proposed GAN-based methods have the great potential of shortening the calibration time and cutting down the budget for various real-world BCI-based applications.
Drowsiness reduces concentration and increases response time, which causes fatal road accidents. Monitoring drivers' drowsiness levels by electroencephalogram (EEG) and taking action may prevent road accidents. EEG signals effectively monitor the driver's mental state as they can monitor brain dynamics. However, calibration is required in advance because EEG signals vary between and within subjects. Because of the inconvenience, calibration has reduced the accessibility of the brain-computer interface (BCI). Developing a generalized classification model is similar to domain generalization, which overcomes the domain shift problem. Especially data augmentation is frequently used. This paper proposes a calibration-free framework for driver drowsiness state classification using manifold-level augmentation. This framework increases the diversity of source domains by utilizing features. We experimented with various augmentation methods to improve the generalization performance. Based on the results of the experiments, we found that deeper models with smaller kernel sizes improved generalizability. In addition, applying an augmentation at the manifold-level resulted in an outstanding improvement. The framework demonstrated the capability for calibration-free BCI.
We present our Brain-Computer Interface (BCI) System, developed for the BCI discipline of Cybathlon 2020 competition. In the BCI discipline, subjects with tetraplegia are required to control a computer game with mental commands. The absolute of the Fast-Fourier Transformation amplitude was calculated as a feature (FFTabs) from one-second-long electroencephalographic (EEG) signals. To extract the final features, we introduced two methods, namely the Feature Average, where the average of the FFTabs for a specific frequency band was calculated, and the Feature Range, which was based on generating multiple Feature Averages for non-overlapping 2 Hz wide frequency bins. The resulting features were fed to a Support Vector Machine classifier. The algorithms were tested on the PhysioNet database and our dataset, which contains 16 offline experiments recorded with 2 tetraplegic subjects. 27 gameplay trials (out of 59) with our tetraplegic subjects reached the 240-second qualification time limit. The Feature Average of canonical frequency bands (alpha, beta, gamma, and theta) were compared with our suggested range30 and range40 methods. On the PhysioNet dataset, the range40 method combined with an ensemble SVM classifier significantly reached the highest accuracy level (0.4607), with a 4-class classification, and outperformed the state-of-the-art EEGNet.
Speeding up the spelling in event-related potentials (ERP) based Brain-Computer Interfaces (BCI) requires eliciting strong brain responses in a short span of time, as much as the accurate classification of such evoked potentials remains challenging and imposes hard constraints for signal processing and machine learning techniques. Recent advances in stimulus presentation and deep learning showcased a promising direction in significantly improving the efficacy of those systems, in this study we propose the combination of colored inverted face stimulation with classification using convolutional neural networks in the hard settings of dry electrodes and fast flashing single-trial ERP-based BCI. The high online accuracy achieved, with two subjects passing the 90 percent correct symbol detection bar and a transfer rate above 60 bits per minute, demonstrates the approach potential in improving the practicality of ERP based BCIs.
Decoding visual stimuli from brain recordings aims to deepen our understanding of the human visual system and build a solid foundation for bridging human and computer vision through the Brain-Computer Interface. However, reconstructing high-quality images with correct semantics from brain recordings is a challenging problem due to the complex underlying representations of brain signals and the scarcity of data annotations. In this work, we present MinD-Vis: Sparse Masked Brain Modeling with Double-Conditioned Latent Diffusion Model for Human Vision Decoding. Firstly, we learn an effective self-supervised representation of fMRI data using mask modeling in a large latent space inspired by the sparse coding of information in the primary visual cortex. Then by augmenting a latent diffusion model with double-conditioning, we show that MinD-Vis can reconstruct highly plausible images with semantically matching details from brain recordings using very few paired annotations. We benchmarked our model qualitatively and quantitatively; the experimental results indicate that our method outperformed state-of-the-art in both semantic mapping (100-way semantic classification) and generation quality (FID) by 66% and 41% respectively. An exhaustive ablation study was also conducted to analyze our framework.
In brain-computer interface or neuroscience applications, generalized canonical correlation analysis (GCCA) is often used to extract correlated signal components in the neural activity of different subjects attending to the same stimulus. This allows quantifying the so-called inter-subject correlation or boosting the signal-to-noise ratio of the stimulus-following brain responses with respect to other (non-)neural activity. GCCA is, however, stimulus-unaware: it does not take the stimulus information into account and does therefore not cope well with lower amounts of data or smaller groups of subjects. We propose a novel stimulus-informed GCCA algorithm based on the MAXVAR-GCCA framework. We show the superiority of the proposed stimulus-informed GCCA method based on the inter-subject correlation between electroencephalography responses of a group of subjects listening to the same speech stimulus, especially for lower amounts of data or smaller groups of subjects.
Electroencephalograms (EEGs) are brain dynamics measured outside the brain, which have been widely utilized in non-invasive brain-computer interface applications. Recently, various neural network approaches have been proposed to improve the accuracy of EEG signal recognition. However, these approaches severely rely on manually designed network structures for different tasks which generally are not sharing the same empirical design cross-task-wise. In this paper, we propose a cross-task neural architecture search (CTNAS-EEG) framework for EEG signal recognition, which can automatically design the network structure across tasks and improve the recognition accuracy of EEG signals. Specifically, a compatible search space for cross-task searching and an efficient constrained searching method is proposed to overcome challenges brought by EEG signals. By unifying structure search on different EEG tasks, this work is the first to explore and analyze the searched structure difference cross-task-wise. Moreover, by introducing architecture search, this work is the first to analyze model performance by customizing model structure for each human subject. Detailed experimental results suggest that the proposed CTNAS-EEG could reach state-of-the-art performance on different EEG tasks, such as Motor Imagery (MI) and Emotion recognition. Extensive experiments and detailed analysis are provided as a good reference for follow-up researchers.
Objective: Steady-state visually evoked potentials (SSVEPs), measured with EEG (electroencephalogram), yield decent information transfer rates (ITR) in brain-computer interface (BCI) spellers. However, the current high performing SSVEP BCI spellers in the literature require an initial lengthy and tiring user-specific training for each new user for system adaptation, including data collection with EEG experiments, algorithm training and calibration (all are before the actual use of the system). This impedes the widespread use of BCIs. To ensure practicality, we propose a highly novel target identification method based on an ensemble of deep neural networks (DNNs), which does not require any sort of user-specific training. Method: We exploit already-existing literature datasets from participants of previously conducted EEG experiments to train a global target identifier DNN first, which is then fine-tuned to each participant. We transfer this ensemble of fine-tuned DNNs to the new user instance, determine the k most representative DNNs according to the participants' statistical similarities to the new user, and predict the target character through a weighted combination of the ensemble predictions. Results: On two large-scale benchmark and BETA datasets, our method achieves impressive 155.51 bits/min and 114.64 bits/min ITRs. Code is available for reproducibility: https://github.com/osmanberke/Ensemble-of-DNNs Conclusion: The proposed method significantly outperforms all the state-of-the-art alternatives for all stimulation durations in [0.2-1.0] seconds on both datasets. Significance: Our Ensemble-DNN method has the potential to promote the practical widespread deployment of BCI spellers in daily lives as we provide the highest performance while enabling the immediate system use without any user-specific training.
This study examines the efficacy of various neural network (NN) models in interpreting mental constructs via electroencephalogram (EEG) signals. Through the assessment of 16 prevalent NN models and their variants across four brain-computer interface (BCI) paradigms, we gauged their information representation capability. Rooted in comprehensive literature review findings, we proposed EEGNeX, a novel, purely ConvNet-based architecture. We pitted it against both existing cutting-edge strategies and the Mother of All BCI Benchmarks (MOABB) involving 11 distinct EEG motor imagination (MI) classification tasks and revealed that EEGNeX surpasses other state-of-the-art methods. Notably, it shows up to 2.1%-8.5% improvement in the classification accuracy in different scenarios with statistical significance (p < 0.05) compared to its competitors. This study not only provides deeper insights into designing efficient NN models for EEG data but also lays groundwork for future explorations into the relationship between bioelectric brain signals and NN architectures. For the benefit of broader scientific collaboration, we have made all benchmark models, including EEGNeX, publicly available at (https://github.com/chenxiachan/EEGNeX).
Electroencephalography (EEG) provides access to neuronal dynamics non-invasively with millisecond resolution, rendering it a viable method in neuroscience and healthcare. However, its utility is limited as current EEG technology does not generalize well across domains (i.e., sessions and subjects) without expensive supervised re-calibration. Contemporary methods cast this transfer learning (TL) problem as a multi-source/-target unsupervised domain adaptation (UDA) problem and address it with deep learning or shallow, Riemannian geometry aware alignment methods. Both directions have, so far, failed to consistently close the performance gap to state-of-the-art domain-specific methods based on tangent space mapping (TSM) on the symmetric positive definite (SPD) manifold. Here, we propose a theory-based machine learning framework that enables, for the first time, learning domain-invariant TSM models in an end-to-end fashion. To achieve this, we propose a new building block for geometric deep learning, which we denote SPD domain-specific momentum batch normalization (SPDDSMBN). A SPDDSMBN layer can transform domain-specific SPD inputs into domain-invariant SPD outputs, and can be readily applied to multi-source/-target and online UDA scenarios. In extensive experiments with 6 diverse EEG brain-computer interface (BCI) datasets, we obtain state-of-the-art performance in inter-session and -subject TL with a simple, intrinsically interpretable network architecture, which we denote TSMNet.
Stable and accurate electroencephalogram (EEG) signal acquisition is fundamental in non-invasive brain-computer interface (BCI) technology. Commonly used EEG acquisition system's hardware and software are usually closed-source. Its inability to flexible expansion and secondary development is a major obstacle to real-time BCI research. This paper presents the Beijing University of Posts and Telecommunications EEG Acquisition Tool System named BEATS. It implements a comprehensive system from hardware to software, composed of the analog front-end, microprocessor, and software platform. BEATS is capable of collecting 32-channel EEG signals at a guaranteed sampling rate of 4k Hz with wireless transmission. Compared to state-of-the-art systems used in many EEG fields, it displays a better sampling rate. Using techniques including direct memory access, first in first out, and timer, the precision and stability of the acquisition are ensured at the microsecond level. An evaluation is conducted during 24 hours of continuous acquisitions. The data loss is 0 packets and the average maximum delay is only 0.07 s/h. Moreover, as an open source system, BEATS provides detailed design files, and adopts a plug-in structure and easy-to-access materials, which makes it can be quickly reproduced. Schematics, source code, and other materials of BEATS are available at https://github.com/buptantEEG/BEATS.
While Artificial Intelligence (AI) technologies are being progressively developed, artists and researchers are investigating their role in artistic practices. In this work, we present an AI-based Brain-Computer Interface (BCI) in which humans and machines interact to express feelings artistically. This system and its production of images give opportunities to reflect on the complexities and range of human emotions and their expressions. In this discussion, we seek to understand the dynamics of this interaction to reach better co-existence in fairness, inclusion, and aesthetics.