Objective: Brain-computer interfaces (BCI) that enables people with severe motor disabilities to use their brain signals for direct control of objects have attracted increased interest in rehabilitation. To date, no study has investigated feasibility of the BCI framework incorporating both intracortical and scalp signals. Methods: Concurrent local field potential (LFP) from the hand-knob area and scalp EEG were recorded in a paraplegic patient undergoing a spike-based close-loop neurorehabilitation training. Based upon multimodal spatio-spectral feature extraction and Naïve Bayes classification, we developed, for the first time, a novel LFP-EEG-BCI for motor intention decoding. A transfer learning (TL) approach was employed to further improve the feasibility. The performance of the proposed LFP-EEG-BCI for four-class upper-limb motor intention decoding was assessed. Results: Using a decision fusion strategy, we showed that the LFP-EEG-BCI significantly (p $< $ 0.05) outperformed single modal BCI (LFP-BCI and EEG-BCI) in terms of decoding accuracy with the best performance achieved using regularized common spatial pattern features. Interrogation of feature characteristics revealed discriminative spatial and spectral patterns, which may lead to new insights for better understanding of brain dynamics during different motor imagery tasks and promote development of efficient decoding algorithms. Moreover, we showed that similar classification performance could be obtained with few training trials, therefore highlighting the efficacy of TL. Conclusion: The present findings demonstrated the superiority of the novel LFP-EEG-BCI in motor intention decoding. Significance: This work introduced a novel LFP-EEG-BCI that may lead to new directions for developing practical neurorehabilitation systems with high detection accuracy and multi-paradigm feasibility in clinical applications.
4 January 2022 in IEEE Journal of Biomedical and Health Informatics
The prediction of schizophrenia-related psychopathologic deficits is exceedingly important in the fields of psychiatry and clinical practice. However, objective association of the brain structure alterations to the illness clinical symptoms is challenging. Although, schizophrenia has been characterized as a brain dysconnectivity syndrome, evidence accounting for neuroanatomical network alterations remain scarce. Moreover, the absence of generalized connectome biomarkers for the assessment of illness progression further perplexes the prediction of long-term symptom severity. In this paper, a combination of individualized prediction models with quantitative graph theoretical analysis was adopted, providing a comprehensive appreciation of the extent to which the brain network properties are affected over time in schizophrenia. Specifically, Connectome-based Prediction Models were employed on Structural Connectivity (SC) features, efficiently capturing individual network-related differences, while identifying the anatomical connectivity disturbances contributing to the prediction of psychopathological deficits. Our results demonstrated distinctions among widespread cortical circuits responsible for different domains of symptoms, indicating the complex neural mechanisms underlying schizophrenia. Furthermore, the generated models were able to significantly predict changes of symptoms using SC features at follow-up, while the preserved SC features suggested an association with improved positive and overall symptoms. Moreover, cross-sectional significant deficits were observed in network efficiency and a progressive aberration of global integration in patients compared to healthy controls, representing a group-consensus pathological map, while supporting the dysconnectivity hypothesis.
13 January 2022 in IEEE Transactions on Neural Systems and Rehabilitation Engineering
Because of the undesired fatigue-related consequences, accumulating efforts have been made to find an effective intervention to alleviate the suboptimal cognitive function caused by mental fatigue. Nonetheless, limitations of intervention and evaluation methods may hinder the revealing of underlying neural mechanisms of fatigue recovery. Through the newly-developed dynamic functional connectivity (FC) analysis framework, this study aims to investigate the effects of two types of mid-task interventions (i.e., rest-break and moderate-intensity exercise-break) on the dynamic reorganization of FC during the execution of psychomotor vigilance test (PVT). Using a sliding window approach, temporal brain networks within each frequency band (i.e., $\delta $ , $\theta $ , $\alpha $ , & $\beta $ ) were estimated before and immediately after the intervention, and towards the end of the task to investigate the immediate and delayed effects respectively during post-break task reengagement. Behaviourally, similar beneficial effects of exercise- and rest-break on performance were observed, manifested by the immediate improvements after both interventions and a long-lasting influence towards the end of tasks. Moreover, temporal brain networks assessment showed significant immediate decreases of fluctuability, which was followed by an increase of fluctuability towards the end of intervention tasks. Furthermore, the temporal nodal measure revealed the channels with significant differences across tasks were mainly resided in the fronto-parietal areas that exhibited interesting frequency-dependent distribution. The observations of immediate and delayed dynamic FC reorganizations extend previous fatigue-related intervention and static FC studies, and provide new insight into the dynamic characteristics of FC during post-break task reengagement.
Accumulating efforts have been made to investigate cognitive impairment in stroke patients, but little has been focused on mild stroke. Research on the impact of mild stroke and different lesion locations on cognitive impairment is still limited. To investigate the underlying mechanisms of cognitive dysfunction in mild stroke at different lesion locations, electroencephalograms (EEGs) were recorded in three groups (40 patients with cortical stroke (CS), 40 patients with subcortical stroke (SS), and 40 healthy controls (HC)) during a visual oddball task. Power envelope connectivity (PEC) was constructed based on EEG source signals, followed by graph theory analysis to quantitatively assess functional brain network properties. A classification framework was further applied to explore the feasibility of PEC in the identification of mild stroke. The results showed worse behavioral performance in the patient groups, and PECs with significant differences among three groups showed complex distribution patterns in frequency bands and the cortex. In the delta band, the global efficiency was significantly higher in HC than in CS (p = 0.011), while local efficiency was significantly increased in SS than in CS (p = 0.038). In the beta band, the small-worldness was significantly increased in HC compared to CS (p = 0.004). Moreover, the satisfactory classification results (76.25% in HC vs. CS, and 80.00% in HC vs. SS) validate the potential of PECs as a biomarker in the detection of mild stroke. Our findings offer some new quantitative insights into the complex mechanisms of cognitive impairment in mild stroke at different lesion locations, which may facilitate post-stroke cognitive rehabilitation.
This study explores the effectiveness of functional near-infrared spectroscopy (fNIRS) as an implicit measurement tool for evaluating sweetness intensity and affective value. Thirty-two participants tasted sucrose solutions at concentrations of 0.15 M, 0.3 M, and 0.6 M, while both their neural responses were recorded with a 24-channel fNIRS system and their self-reported assessments of sweetness intensity and affective value were collected. The neural fNIRS data were converted into oxygenated hemoglobin (HbO) and deoxygenated hemoglobin (HbR) concentrations using the modified Beer–Lambert Law, and analyzed through univariate activation analysis and multivariable decoding analysis to identify neural activation patterns associated with sweetness perception. The results showed significant activation in the dorsolateral prefrontal cortex (dlPFC) and orbitofrontal cortex (OFC) in response to varying levels of sweetness intensity and affective value, with channels 8, 10, 12, 13, 14, 15, and 17 consistently activated across all sucrose concentrations. As sweetness concentration increased from 0.15 M to 0.6 M, the number of significantly activated channels rose from seven to eleven, indicating stronger and more widespread neural responses corresponding to higher sweetness intensity. The multivariable decoding analysis further demonstrated the capability of fNIRS in accurately distinguishing positive affective responses, with up to 72.1% accuracy. The moderate positive correlation between explicit self-reports and implicit fNIRS data regarding sweetness intensity further supports the validity of fNIRS as a reliable tool for assessing taste perception. This study highlights the potential of fNIRS in sensory neuroscience, demonstrating its effectiveness in capturing the neural mechanisms underlying sweet taste perception.
Recently multimodal learning has become a popular research topic because variety of types of sensors can be used to describe the events, for example, images and texts in the social network for disaster detection and response. It is useful because of different sources of information that data will cause heterogeneous problems. The kernel function can measure the similarity of data, and the fusion of different kernels can handle this problem well. This paper proposes a method based on multimodal deep kernel learning to detect specific categories of content posted on Twitter during a disaster for humanitarian response. Specifically, the convolutional neural network is used to extract image and text features, and obtain multiple kernel functions in each mode. Then a deep neural network is used to fuse multiple basic kernel functions corresponding to the two modalities. Finally, we use Support Vector Machine (SVM) for classification. Applying this method to a real disaster data set, and compared with the baseline, the results prove the effectiveness of our model.