Articles by Thinh Nguyen in JoVE
Concurrent EEG and Functional MRI Recording and Integration Analysis for Dynamic Cortical Activity Imaging Thinh Nguyen*1, Thomas Potter*1, Christof Karmonik2, Robert Grossman2, Yingchun Zhang1,3 1Department of Biomedical Engineering, Cullen College of Engineering, University of Houston, 2Department of Neurosurgery, Houston Methodist Hospital and Research Institute, 3Guangdong Provincial Work Injury Rehabilitation Center An EEG-fMRI multimodal imaging method, known as the spatiotemporal fMRI-constrained EEG source imaging method, is described here. The presented method employs conditionally-active fMRI sub-maps, or priors, to guide EEG source localization in a manner that improves spatial specificity and limits erroneous results.
Other articles by Thinh Nguyen on PubMed
EEG Source Imaging Guided by Spatiotemporal Specific FMRI: Toward an Understanding of Dynamic Cognitive Processes Neural Plasticity. 2016 | Pubmed ID: 27803816 Understanding the mechanism of neuroplasticity is the first step in treating neuromuscular system impairments with cognitive rehabilitation approaches. To characterize the dynamics of the neural networks and the underlying neuroplasticity of the central motor system, neuroimaging tools with high spatial and temporal accuracy are desirable. EEG and fMRI stand among the most popular noninvasive neuroimaging modalities with complementary features, yet achieving both high spatial and temporal accuracy remains a challenge. A novel multimodal EEG/fMRI integration method was developed in this study to achieve high spatiotemporal accuracy by employing the most probable fMRI spatial subsets to guide EEG source localization in a time-variant fashion. In comparison with the traditional fMRI constrained EEG source imaging method in a visual/motor activation task study, the proposed method demonstrated superior localization accuracy with lower variation and identified neural activity patterns that agreed well with previous studies. This spatiotemporal fMRI constrained source imaging method was then implemented in a "sequential multievent-related potential" paradigm where motor activation is evoked by emotion-related visual stimuli. Results demonstrate that the proposed method can be used as a powerful neuroimaging tool to unveil the dynamics and neural networks associated with the central motor system, providing insights into neuroplasticity modulation mechanism.
Characterization of Volume-Based Changes in Cortical Auditory Evoked Potentials and Prepulse Inhibition Scientific Reports. Sep, 2017 | Pubmed ID: 28894145 The auditory evoked startle reflex is a conserved response resulting in neurological and motor activity. The presence of a mild prepulse immediately before the main pulse inhibits startle responses, though the mechanism for this remains unknown. In this study, the electroencephalography (EEG) data recorded from 15 subjects was analyzed to study the N1 and P2 components of cortical auditory evoked potentials (CAEPs) evoked by 70, 80, 90, 100, and 110 dB stimuli both in the presence and absence of 70 dB prepulses. Results without a prepulse showed an evolution of N1 amplitudes, increasing with stimulus intensity and showing largely significant differences. Results from prepulse trials only showed noteworthy changes in peak-to-peak amplitude in the 100 dB condition. Prepulse and non-prepulse conditions were then compared using peak amplitudes and theta power. Prepulse conditions significantly decreased the amplitude for both components in the 110 dB condition, i.e., pre-pulse inhibition, but significantly increased the N1 amplitude in the 70 dB condition, i.e., pre-pulse facilitation. Similarly theta band power significantly increased in the 70 dB prepulse condition and significantly decreased in the 110 dB prepulse condition. These results expand the basis of knowledge regarding how CAEPs change and elaborate on their neural function and representation.
Characterization of Dynamic Changes of Current Source Localization Based on Spatiotemporal FMRI Constrained EEG Source Imaging Journal of Neural Engineering. Jun, 2018 | Pubmed ID: 29214978 Neuroimaging has been employed as a promising approach to advance our understanding of brain networks in both basic and clinical neuroscience. Electroencephalography (EEG) and functional magnetic resonance imaging (fMRI) represent two neuroimaging modalities with complementary features; EEG has high temporal resolution and low spatial resolution while fMRI has high spatial resolution and low temporal resolution. Multimodal EEG inverse methods have attempted to capitalize on these properties but have been subjected to localization error. The dynamic brain transition network (DBTN) approach, a spatiotemporal fMRI constrained EEG source imaging method, has recently been developed to address these issues by solving the EEG inverse problem in a Bayesian framework, utilizing fMRI priors in a spatial and temporal variant manner. This paper presents a computer simulation study to provide a detailed characterization of the spatial and temporal accuracy of the DBTN method.