Byoung-Kyong Min | Korea University, Republic of Korea (original) (raw)

Uploads

Papers by Byoung-Kyong Min

Mental workload is defined as the proportion of the information processing capability used to per... more Mental workload is defined as the proportion of the information processing capability used to perform a task. High cognitive load requires additional resources that may reduce the processing efficiency and performance. Therefore, the technique of workload estimation can ensure a proper working environment to promote the working efficiency of each person. In this paper, we propose a three-dimensional convolutional neural network (3D CNN) based a multilevel feature fusion algorithm for mental workload estimation using electroencephalogram (EEG) signals. Raw EEG signals are converted to 3D EEG images and then multilevel features are extracted in each layer of the 3D convolution operation and each multilevel feature is multiplied by a weighting factor, which determines the importance of the feature. The accuracy of our network is 90.3%, which is better than conventional algorithms.

Trends in Biotechnology, Sep 1, 2020

Most of the studies employing neuroimaging have focused on cortical and subcortical signals indiv... more Most of the studies employing neuroimaging have focused on cortical and subcortical signals individually to obtain neurophysiological signatures of cognitive functions. However, understanding the dynamic communication between the cortex and subcortical structures is essential for unraveling the neural correlates of cognition. In this quest, magnetoencephalography (MEG) and electroencephalography (EEG) are the methods of choice because they are non-invasive electrophysiological recording techniques with high temporal resolution. Sophisticated MEG/EEG source estimation techniques and network analysis methods developed recently can provide a more comprehensive understanding of the neurophysiological mechanisms of fundamental cognitive processes. Used together with noninvasive modulation of cortical-subcortical communication, these approaches may open up new possibilities for expanding the repertoire of non-invasive cognitive neurotechnology.

NeuroImage, Nov 1, 2010

People often fail to notice a large change in the visual scene when the change occurs during a br... more People often fail to notice a large change in the visual scene when the change occurs during a brief interruption of the viewing. Since the change is well above perceptual threshold in continuous viewing, the failure (termed change blindness) has been attributed to abnormal visual short-term memory (VSTM). However, it is still unclear where the abnormality lies among the phases in VSTM, namely, encoding, maintenance, and retrieval-comparison. EEG oscillations, especially the gamma activity, have been suggested as neural signatures of VSTM, but have not been examined in the context of change blindness. Thus, we asked in the present study whether change detection or failure is correlated with EEG oscillatory activities and, if so, whether the timing and the spatial distribution of the oscillations could pin-point the abnormal phase of VSTM in change blindness. While on EEG recording, subjects watched morphed pictures of human faces in trials which consisted of a 200-ms initial image display, a 500-ms blank period, and a 200-ms comparison image display. The two images were either the same or clearly different above threshold. Trials with different images were classified as hit or missed, based on subjects' responses, and EEG data were compared between the two types of trials. Enhanced gamma activity was observed in the right temporal-parietal region during all periods in the hit trials compared to the missed ones. Frontal theta activity was increased during initial image encoding, whereas beta activity was decreased during maintenance and retrieval-comparison in the hit trials. These results point to weak encoding of initial images as the culprit for a later failure in change detection, while abnormal processing in subsequent phases of VSTM may result from the weak encoding and also contribute to change blindness.

Ultrasound in Medicine and Biology, 2015

Brain machine interfaces (BMIs) enable us to control external devices using our brain signals. Us... more Brain machine interfaces (BMIs) enable us to control external devices using our brain signals. Using a grid-shaped flickering line-array and a shrink-rLDA classifier, top-down information could recently be decoded in a steady-state visual evoked potential (SSVEP)-based BMI paradigm. The present study tested its feasibility in online implementation. We found that within reasonable computing time (0.114 s on average) its online system was successfully accomplished with a decoding accuracy of 53.7% on average. The accuracy was 3.2 times significantly higher than the accuracy by random-shuffled data (16.7%). Therefore, using the grid-shaped SSVEP-based BMI, one's multiclass (at least 6 classes) intention can be online decoded and subsequently control external devices.

Scientific Reports, Nov 3, 2016

Brain Research, Nov 1, 2013

Electroencephalography (EEG) has become a popular tool in brain-computer interface (BCI) research... more Electroencephalography (EEG) has become a popular tool in brain-computer interface (BCI) research. Some of the drawbacks pertain to the offline analyses of the neural signal that prevent the subjects from engaging in real-time error correction during learning. Other limitations include the complex nature of the visual stimuli, often inducing fatigue and introducing considerable delays, possibly interfering with spontaneous performance. By replacing the complex external visual input with internally driven motor imagery we can overcome some delay problems, at the expense of losing the ability to precisely parameterize features of the input stimulus. To address these issues we here introduce a direction-imagery task to BCI. We observed that all participants showed almost perfect performance in the fourth session. Participants reported that as they mastered the mental control with direct thinking of direction. These observations provide corroborative evidence for practicability of prefrontal signals to be used as promising cognitive BCI commands.

Trends in Biotechnology, Jul 1, 2014

Frontiers in Human Neuroscience, Apr 26, 2023

Brain-machine interfaces (BMIs) enable humans to control devices by modulating their brain signal... more Brain-machine interfaces (BMIs) enable humans to control devices by modulating their brain signals. As the current BMI technology has several obstacles to overcome, additional sources of brain activity need to be explored. It seems plausible that the brain activity associated with top-down cognitive functions could open a new prospect in the field of BMIs. As top-down cognitive BMIs could exploit neural signals from more diverse networks, a deep-learning approach with complex hidden layers may provide a more optimal decoding performance. In this study, using our top-down steady-state visual evoked potential (SSVEP) paradigm (N = 20), we observed that the decoding accuracy (48.42%) of a deep-learning algorithm with a sigmoid activation function was significantly higher than that of regularized linear discriminant analysis (rLDA) with shrinkage (42.52%; t(19) = −3.183, p < 0.01), used in our previous study. Therefore, a deep-learning approach seems to be more optimized for classification in the top-down cognitive BMI paradigm.

Trends in Biotechnology, Jul 1, 2017

NeuroImage, Aug 1, 2021

The prefrontal cortex (PFC) plays a pivotal role in goal-directed cognition, yet its representati... more The prefrontal cortex (PFC) plays a pivotal role in goal-directed cognition, yet its representational code remains an open problem with decoding techniques ineffective in disentangling task-relevant variables from PFC. Here we applied regularized linear discriminant analysis to human scalp EEG data and were able to distinguish a mental-rotation task versus a color-perception task with 87% decoding accuracy. Dorsal and ventral areas in lateral PFC provided the dominant features dissociating the two tasks. Our findings show that EEG can reliably decode two independent task states from PFC and emphasize the PFC dorsal/ventral functional specificity in processing the where rotation task versus the what color task.

Theoretical Biology and Medical Modelling, Mar 30, 2010

Mental workload is defined as the proportion of the information processing capability used to per... more Mental workload is defined as the proportion of the information processing capability used to perform a task. High cognitive load requires additional resources that may reduce the processing efficiency and performance. Therefore, the technique of workload estimation can ensure a proper working environment to promote the working efficiency of each person. In this paper, we propose a three-dimensional convolutional neural network (3D CNN) based a multilevel feature fusion algorithm for mental workload estimation using electroencephalogram (EEG) signals. Raw EEG signals are converted to 3D EEG images and then multilevel features are extracted in each layer of the 3D convolution operation and each multilevel feature is multiplied by a weighting factor, which determines the importance of the feature. The accuracy of our network is 90.3%, which is better than conventional algorithms.

Trends in Biotechnology, Sep 1, 2020

Most of the studies employing neuroimaging have focused on cortical and subcortical signals indiv... more Most of the studies employing neuroimaging have focused on cortical and subcortical signals individually to obtain neurophysiological signatures of cognitive functions. However, understanding the dynamic communication between the cortex and subcortical structures is essential for unraveling the neural correlates of cognition. In this quest, magnetoencephalography (MEG) and electroencephalography (EEG) are the methods of choice because they are non-invasive electrophysiological recording techniques with high temporal resolution. Sophisticated MEG/EEG source estimation techniques and network analysis methods developed recently can provide a more comprehensive understanding of the neurophysiological mechanisms of fundamental cognitive processes. Used together with noninvasive modulation of cortical-subcortical communication, these approaches may open up new possibilities for expanding the repertoire of non-invasive cognitive neurotechnology.

NeuroImage, Nov 1, 2010

People often fail to notice a large change in the visual scene when the change occurs during a br... more People often fail to notice a large change in the visual scene when the change occurs during a brief interruption of the viewing. Since the change is well above perceptual threshold in continuous viewing, the failure (termed change blindness) has been attributed to abnormal visual short-term memory (VSTM). However, it is still unclear where the abnormality lies among the phases in VSTM, namely, encoding, maintenance, and retrieval-comparison. EEG oscillations, especially the gamma activity, have been suggested as neural signatures of VSTM, but have not been examined in the context of change blindness. Thus, we asked in the present study whether change detection or failure is correlated with EEG oscillatory activities and, if so, whether the timing and the spatial distribution of the oscillations could pin-point the abnormal phase of VSTM in change blindness. While on EEG recording, subjects watched morphed pictures of human faces in trials which consisted of a 200-ms initial image display, a 500-ms blank period, and a 200-ms comparison image display. The two images were either the same or clearly different above threshold. Trials with different images were classified as hit or missed, based on subjects&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;#39; responses, and EEG data were compared between the two types of trials. Enhanced gamma activity was observed in the right temporal-parietal region during all periods in the hit trials compared to the missed ones. Frontal theta activity was increased during initial image encoding, whereas beta activity was decreased during maintenance and retrieval-comparison in the hit trials. These results point to weak encoding of initial images as the culprit for a later failure in change detection, while abnormal processing in subsequent phases of VSTM may result from the weak encoding and also contribute to change blindness.

Ultrasound in Medicine and Biology, 2015

Brain machine interfaces (BMIs) enable us to control external devices using our brain signals. Us... more Brain machine interfaces (BMIs) enable us to control external devices using our brain signals. Using a grid-shaped flickering line-array and a shrink-rLDA classifier, top-down information could recently be decoded in a steady-state visual evoked potential (SSVEP)-based BMI paradigm. The present study tested its feasibility in online implementation. We found that within reasonable computing time (0.114 s on average) its online system was successfully accomplished with a decoding accuracy of 53.7% on average. The accuracy was 3.2 times significantly higher than the accuracy by random-shuffled data (16.7%). Therefore, using the grid-shaped SSVEP-based BMI, one's multiclass (at least 6 classes) intention can be online decoded and subsequently control external devices.

Scientific Reports, Nov 3, 2016

Brain Research, Nov 1, 2013

Electroencephalography (EEG) has become a popular tool in brain-computer interface (BCI) research... more Electroencephalography (EEG) has become a popular tool in brain-computer interface (BCI) research. Some of the drawbacks pertain to the offline analyses of the neural signal that prevent the subjects from engaging in real-time error correction during learning. Other limitations include the complex nature of the visual stimuli, often inducing fatigue and introducing considerable delays, possibly interfering with spontaneous performance. By replacing the complex external visual input with internally driven motor imagery we can overcome some delay problems, at the expense of losing the ability to precisely parameterize features of the input stimulus. To address these issues we here introduce a direction-imagery task to BCI. We observed that all participants showed almost perfect performance in the fourth session. Participants reported that as they mastered the mental control with direct thinking of direction. These observations provide corroborative evidence for practicability of prefrontal signals to be used as promising cognitive BCI commands.

Trends in Biotechnology, Jul 1, 2014

Frontiers in Human Neuroscience, Apr 26, 2023

Brain-machine interfaces (BMIs) enable humans to control devices by modulating their brain signal... more Brain-machine interfaces (BMIs) enable humans to control devices by modulating their brain signals. As the current BMI technology has several obstacles to overcome, additional sources of brain activity need to be explored. It seems plausible that the brain activity associated with top-down cognitive functions could open a new prospect in the field of BMIs. As top-down cognitive BMIs could exploit neural signals from more diverse networks, a deep-learning approach with complex hidden layers may provide a more optimal decoding performance. In this study, using our top-down steady-state visual evoked potential (SSVEP) paradigm (N = 20), we observed that the decoding accuracy (48.42%) of a deep-learning algorithm with a sigmoid activation function was significantly higher than that of regularized linear discriminant analysis (rLDA) with shrinkage (42.52%; t(19) = −3.183, p < 0.01), used in our previous study. Therefore, a deep-learning approach seems to be more optimized for classification in the top-down cognitive BMI paradigm.

Trends in Biotechnology, Jul 1, 2017

NeuroImage, Aug 1, 2021

The prefrontal cortex (PFC) plays a pivotal role in goal-directed cognition, yet its representati... more The prefrontal cortex (PFC) plays a pivotal role in goal-directed cognition, yet its representational code remains an open problem with decoding techniques ineffective in disentangling task-relevant variables from PFC. Here we applied regularized linear discriminant analysis to human scalp EEG data and were able to distinguish a mental-rotation task versus a color-perception task with 87% decoding accuracy. Dorsal and ventral areas in lateral PFC provided the dominant features dissociating the two tasks. Our findings show that EEG can reliably decode two independent task states from PFC and emphasize the PFC dorsal/ventral functional specificity in processing the where rotation task versus the what color task.

Theoretical Biology and Medical Modelling, Mar 30, 2010