Objective: Functional near-infrared spectroscopy (fNIRS) has recently gained momentum in research on motor-imagery (MI)-based brain-computer interfaces (BCIs). However, strikingly, most of the research effort is primarily devoted to enhancing fNIRS-based BCIs for healthy individuals. The ability of patients with amyotrophic lateral sclerosis (ALS), among the main BCI end-users to utilize fNIRS-based hemodynamic responses to efficiently control an MI-based BCI, has not yet been explored. This study aims to quantify subject-specific spatio-temporal characteristics of ALS patients’ hemodynamic responses to MI tasks, and to investigate the feasibility of using these responses as a means of communication to control a binary BCI. Methods: Hemodynamic responses were recorded using fNIRS from eight patients with ALS while performing MI-Rest tasks. The generalized linear model (GLM) analysis was conducted to statistically estimate and evaluate individualized spatial activation. Selected channel sets were statistically optimized for classification. Subject-specific discriminative features, including a proposed data-driven estimated coefficient obtained from GLM, and optimized classification parameters were identified and used to further evaluate the performance using a linear support vector machine (SVM) classifier. Results: Inter-subject variations were observed in spatio-temporal characteristics of patients’ hemodynamic responses. Using optimized classification parameters and feature sets, all subjects could successfully use their MI hemodynamic responses to control a BCI with an average classification accuracy of 85.4%±19.8%. Significance: Our results indicate a promising application of fNIRS-based MI hemodynamic responses to control a binary BCI by ALS patients. These findings highlight the importance of subject-specific data-driven approaches for identifying discriminative spatio-temporal characteristics for an optimized BCI performance.

Objective: Amyotrophic lateral sclerosis (ALS) is a complex neurodegenerative disease that causes the progressive loss of voluntary muscle control. Recent studies have reported conflicting results on alterations in resting-state functional brain networks in ALS by adopting unimodal techniques that measure either electrophysiological or vascular-hemodynamic neural functions. However, no study to date has explored simultaneous electrical and vascular-hemodynamic changes in the resting-state brain in ALS. Using complementary multimodal electroencephalography (EEG) and functional near-infrared spectroscopy (fNIRS) recording and analysis techniques, we explored the underlying multidimensional neural contributions to altered oscillations and functional connectivity in people with ALS. Methods: 10 ALS patients and 9 age-matched controls underwent multimodal EEG-fNIRS recording in the resting state. Resting-state functional connectivity (RSFC) and power spectra of both modalities in both groups were analyzed and compared statistically. Results: Increased fronto-parietal EEG connectivity in the alpha and beta bands and increased interhemispheric and right intra-hemispheric fNIRS connectivity in the frontal and prefrontal regions were observed in ALS. Frontal, central, and temporal theta and alpha EEG power decreased in ALS, as did parietal and occipital alpha EEG power, while frontal and parietal hemodynamic spectral power increased in ALS. Significance: These results suggest that electro-vascular disruption in neuronal networks extends to the extra-motor regions in ALS patients, which can ultimately introduce novel neural markers of ALS that can be exploited further as diagnostic and prognostic tools.

The use of electroencephalogram (EEG) or "brain waves" for human-computer interaction is a new and challenging field that has gained momentum in the past few years. If several mental states can be reliably distinguished by recognizing patterns in EEG, then a paralyzed person could communicate to a device like a wheelchair by composing sequences of these mental states. In this research, EEG from one subject who performed three mental tasks have been classified using radial basis function (RBF) support vector machines (SVM) to control overfitting. A method for EEG preprocessing based on independent component analysis (ICA) was proposed and three different feature extraction techniques were compared: parametric autoregressive (AR) modeling, AR spectral analysis and power differences between four frequency bands. The best classification accuracy was approximately 70% using the parametric AR model representation with almost 5% improvement of accuracy over unprocessed data.

Applications of multimodal neuroimaging techniques, including Electroencephalography (EEG) and near-infrared spectroscopy (fNIRS) have gained prominence in recent years, and they are widely practiced in brain-computer interface (BCI) and neuro-pathological diagnosis applications. Most existing approaches assume observations are independent and identically distributed (i.i.d.), as shown in the top section of the right figure, yet ignore the difference among subjects. It has been challenging to model subject groups to maintain topological information (e.g., patient graphs) while fusing BCI signals for discriminant feature learning. In this paper, we introduce a Topology-aware Graph-based Multimodal Fusion (TaGMF) framework to classify amyotrophic lateral sclerosis (ALS) and healthy subjects, illustrated in the lower section of the right image. Our framework is built on graph neural networks (GNN) but with two unique contributions. First, a novel Topology-aware Graph (TaG) is proposed to model subject groups by considering (1) inter-subject, (2) intra-subject, and (3) inter-group relations. Second, the learned representation of EEG and fNIRS signals of each subject allows for explorations of different fusion strategies along with the TaGMF optimizations. Our analysis demonstrates the effectiveness of our graph-based fusion approach in multimodal classification by achieving a 22.6% performance improvement over classical approaches.

The prospect of electroencephalography (EEG) and functional near-infrared spectroscopy (fNIRS) in the presence of topological information of participants is often left unexplored in most of the brain-computer interface (BCI) systems. Additionally, the usage of these modalities together in the field of multimodality analysis to support multiple brain signals toward improving BCI performance is not fully examined. This study first presents a multimodal data fusion framework to exploit and decode the complementary synergistic properties in multimodal neural signals. Moreover, the relations among different subjects and their observations also play critical roles in classifying unknown subjects. We developed a context-aware graph neural network (GNN) model utilizing the pairwise relationship among participants to investigate the performance on an auditory task classification. We explored standard and deviant auditory EEG and fNIRS data where each subject was asked to perform an auditory oddball task and has multiple trials regarded as context-aware nodes in our graph construction. In experiments, our multimodal data fusion strategy showed an improvement up to 8.40% via SVM and 2.02% via GNN, compared to the single-modal EEG or fNIRS. In addition, our context-aware GNN achieved 5.3%, 4.07% and 4.53% higher accuracy for EEG, fNIRS and multimodal data based experiments, compared to the baseline models.

Recent work has demonstrated that P300-based BCI systems can provide long-term communication for individuals with amyotrophic lateral sclerosis (ALS). However, even individuals with a successful history of BCI home use, can experience substantial variation in their day-to-day P300-based BCI performance. Recent studies suggest that functional connectivity among frontal-parietal cortical network governs the generation of P300 components. This study aims to characterize the correlates of EEG spectral and functional connectivity that underlie these fluctuations in performance during use of the P300-based BCI speller at home. Our results suggest that increased frontal-parietal functional connectivity and frontal non-phase locked theta band power modulations are associated with unsuccessful performance. The identified neural patterns suggest possible conscious/unconscious error-related adaptation mechanisms that can be associated with the realization of uncertainty and response conflict. These findings provide new insights into the underlying mechanisms of BCI performance variation and can inform the development of more robust BCI systems.