Sung-Jin Kim

Sung-Jin Kim, Heon-Gyu Kwak, Hyeon-Taek Han et al.

Brain-computer interface (BCI) has garnered the significant attention for their potential in various applications, with event-related potential (ERP) performing a considerable role in BCI systems. This paper introduces a novel Distributed Inference System tailored for detecting task-wise single-trial ERPs in a stream of satellite images. Unlike traditional methodologies that employ a single model for target detection, our system utilizes multiple models, each optimized for specific tasks, ensuring enhanced performance across varying image transition times and target onset times. Our experiments, conducted on four participants, employed two paradigms: the Normal paradigm and an AI paradigm with bounding boxes. Results indicate that our proposed system outperforms the conventional methods in both paradigms, achieving the highest $F_β$ scores. Furthermore, including bounding boxes in the AI paradigm significantly improved target recognition. This study underscores the potential of our Distributed Inference System in advancing the field of ERP detection in satellite image streams.

Sung-Jin Kim, Dae-Hyeok Lee, Yeon-Woo Choi

Brain-computer interface (BCI) is a communication system between humans and computers reflecting human intention without using a physical control device. Since deep learning is robust in extracting features from data, research on decoding electroencephalograms by applying deep learning has progressed in the BCI domain. However, the application of deep learning in the BCI domain has issues with a lack of data and overconfidence. To solve these issues, we proposed a novel data augmentation method, CropCat. CropCat consists of two versions, CropCat-spatial and CropCat-temporal. We designed our method by concatenating the cropped data after cropping the data, which have different labels in spatial and temporal axes. In addition, we adjusted the label based on the ratio of cropped length. As a result, the generated data from our proposed method assisted in revising the ambiguous decision boundary into apparent caused by a lack of data. Due to the effectiveness of the proposed method, the performance of the four EEG signal decoding models is improved in two motor imagery public datasets compared to when the proposed method is not applied. Hence, we demonstrate that generated data by CropCat smooths the feature distribution of EEG signals when training the model.

Dae-Hyeok Lee, Sung-Jin Kim, Yeon-Woo Choi

Non-invasive brain-computer interface technology has been developed for detecting human mental states with high performances. Detection of the pilots' mental states is particularly critical because their abnormal mental states could cause catastrophic accidents. In this study, we presented the feasibility of classifying distraction levels (namely, normal state, low distraction, and high distraction) by applying the deep learning method. To the best of our knowledge, this study is the first attempt to classify distraction levels under a flight environment. We proposed a model for classifying distraction levels. A total of ten pilots conducted the experiment in a simulated flight environment. The grand-average accuracy was 0.8437 for classifying distraction levels across all subjects. Hence, we believe that it will contribute significantly to autonomous driving or flight based on artificial intelligence technology in the future.

Dae-Hyeok Lee, Sung-Jin Kim, Si-Hyun Kim et al.

The detection of pilots' mental states is important due to the potential for their abnormal mental states to result in catastrophic accidents. This study introduces the feasibility of employing deep learning techniques to classify different workload levels, specifically normal state, low workload, and high workload. To the best of our knowledge, this study is the first attempt to classify workload levels of pilots. Our approach involves the hybrid deep neural network that consists of five convolutional blocks and one long short-term memory block to extract the significant features from electroencephalography signals. Ten pilots participated in the experiment, which was conducted within the simulated flight environment. In contrast to four conventional models, our proposed model achieved a superior grand--average accuracy of 0.8613, surpassing other conventional models by at least 0.0597 in classifying workload levels across all participants. Our model not only successfully classified workload levels but also provided valuable feedback to the participants. Hence, we anticipate that our study will make the significant contributions to the advancement of autonomous flight and driving leveraging artificial intelligence technology in the future.

Heejo Kong, Sung-Jin Kim, Gunho Jung et al.

Conventional semi-supervised learning (SSL) ideally assumes that labeled and unlabeled data share an identical class distribution, however in practice, this assumption is easily violated, as unlabeled data often includes unknown class data, i.e., outliers. The outliers are treated as noise, considerably degrading the performance of SSL models. To address this drawback, we propose a novel framework, Diversify and Conquer (DAC), to enhance SSL robustness in the context of open-set semi-supervised learning. In particular, we note that existing open-set SSL methods rely on prediction discrepancies between inliers and outliers from a single model trained on labeled data. This approach can be easily failed when the labeled data is insufficient, leading to performance degradation that is worse than naive SSL that do not account for outliers. In contrast, our approach exploits prediction disagreements among multiple models that are differently biased towards the unlabeled distribution. By leveraging the discrepancies arising from training on unlabeled data, our method enables robust outlier detection even when the labeled data is underspecified. Our key contribution is constructing a collection of differently biased models through a single training process. By encouraging divergent heads to be differently biased towards outliers while making consistent predictions for inliers, we exploit the disagreement among these heads as a measure to identify unknown concepts. Our code is available at https://github.com/heejokong/DivCon.

Dae-Hyeok Lee, Sung-Jin Kim, Si-Hyun Kim

The detection of pilots' mental states is critical, as abnormal mental states have the potential to cause catastrophic accidents. This study demonstrates the feasibility of using deep learning techniques to classify different fatigue levels, specifically a normal state, low fatigue, and high fatigue. To the best of our knowledge, this is the first study to classify fatigue levels in pilots. Our approach employs the hybrid deep neural network comprising five convolutional blocks and one long short-term memory block to extract the significant features from electroencephalography signals. Ten pilots participated in the experiment, which was conducted in a simulated flight environment. Compared to four conventional models, our proposed model achieved a superior grand-average accuracy of 0.8801, outperforming other models by at least 0.0599 in classifying fatigue levels. In addition to successfully classifying fatigue levels, our model provided valuable feedback to subjects. Therefore, we anticipate that our study will make the significant contributions to the advancement of autonomous flight and driving technologies, leveraging artificial intelligence in the future.

Si-Hyun Kim, Sung-Jin Kim, Dae-Hyeok Lee

Brain-computer interface (BCI) enables direct communication between the brain and external devices by decoding neural signals, offering potential solutions for individuals with motor impairments. This study explores the neural signatures of motor execution (ME) and motor imagery (MI) tasks using EEG signals, focusing on four conditions categorized as sense-related (hot and cold) and motor-related (pull and push) conditions. We conducted scalp topography analysis to examine activation patterns in the sensorimotor cortex, revealing distinct regional differences: sense--related conditions primarily activated the posterior region of the sensorimotor cortex, while motor--related conditions activated the anterior region of the sensorimotor cortex. These spatial distinctions align with neurophysiological principles, suggesting condition-specific functional subdivisions within the sensorimotor cortex. We further evaluated the performances of three neural network models-EEGNet, ShallowConvNet, and DeepConvNet-demonstrating that ME tasks achieved higher classification accuracies compared to MI tasks. Specifically, in sense-related conditions, the highest accuracy was observed in the cold condition. In motor-related conditions, the pull condition showed the highest performance, with DeepConvNet yielding the highest results. These findings provide insights into optimizing BCI applications by leveraging specific condition-induced neural activations.

Sung-Jin Kim, Dae-Hyeok Lee, Hyeon-Taek Han

Electroencephalography (EEG) is a generally used neuroimaging approach in brain-computer interfaces due to its non-invasive characteristics and convenience, making it an effective tool for understanding human intentions. Therefore, recent research has focused on decoding human intentions from EEG signals utilizing deep learning methods. However, since EEG signals are highly susceptible to noise during acquisition, there is a high possibility of the existence of noisy data in the dataset. Although pioneer studies have generally assumed that the dataset is well-curated, this assumption is not always met in the EEG dataset. In this paper, we addressed this issue by designing a dataset refinement algorithm that can eliminate noisy data based on metrics evaluating data influence during the training process. We applied the proposed algorithm to two motor imagery EEG public datasets and three different models to perform dataset refinement. The results indicated that retraining the model with the refined dataset consistently led to better generalization performance compared to using the original dataset. Hence, we demonstrated that removing noisy data from the training dataset alone can effectively improve the generalization performance of deep learning models in the EEG domain.

Dae-Hyeok Lee, Sung-Jin Kim, Keon-Woo Lee

Brain-computer interface (BCI) is the technology that enables the communication between humans and devices by reflecting status and intentions of humans. When conducting imagined speech, the users imagine the pronunciation as if actually speaking. In the case of decoding imagined speech-based EEG signals, complex task can be conducted more intuitively, but decoding performance is lower than that of other BCI paradigms. We modified our previous model for decoding imagined speech-based EEG signals. Ten subjects participated in the experiment. The average accuracy of our proposed method was 0.5648 for classifying four words. In other words, our proposed method has significant strength in learning local features. Hence, we demonstrated the feasibility of decoding imagined speech-based EEG signals with robust performance.

Dae-Hyeok Lee, Sung-Jin Kim, Seong-Whan Lee

Brain-computer interface (BCI) is one of the tools which enables the communication between humans and devices by reflecting intention and status of humans. With the development of artificial intelligence, the interest in communication between humans and drones using electroencephalogram (EEG) is increased. Especially, in the case of controlling drone swarms such as direction or formation, there are many advantages compared with controlling a drone unit. Imagined speech is one of the endogenous BCI paradigms, which can identify intentions of users. When conducting imagined speech, the users imagine the pronunciation as if actually speaking. In contrast, overt speech is a task in which the users directly pronounce the words. When controlling drone swarms using imagined speech, complex commands can be delivered more intuitively, but decoding performance is lower than that of other endogenous BCI paradigms. We proposed the Deep-autoleaner (DAL) to learn EEG features of overt speech for imagined speech-based EEG signals classification. To the best of our knowledge, this study is the first attempt to use EEG features of overt speech to decode imagined speech-based EEG signals with an autoencoder. A total of eight subjects participated in the experiment. When classifying four words, the average accuracy of the DAL was 48.41%. In addition, when comparing the performance between w/o and w/ EEG features of overt speech, there was a performance improvement of 7.42% when including EEG features of overt speech. Hence, we demonstrated that EEG features of overt speech could improve the decoding performance of imagined speech.

Dae-Hyeok Lee, Dong-Kyun Han, Sung-Jin Kim et al.

Brain-computer interface (BCI) is used for communication between humans and devices by recognizing status and intention of humans. Communication between humans and a drone using electroencephalogram (EEG) signals is one of the most challenging issues in the BCI domain. In particular, the control of drone swarms (the direction and formation) has more advantages compared to the control of a drone. The visual imagery (VI) paradigm is that subjects visually imagine specific objects or scenes. Reduction of the variability among EEG signals of subjects is essential for practical BCI-based systems. In this study, we proposed the subepoch-wise feature encoder (SEFE) to improve the performances in the subject-independent tasks by using the VI dataset. This study is the first attempt to demonstrate the possibility of generalization among subjects in the VI-based BCI. We used the leave-one-subject-out cross-validation for evaluating the performances. We obtained higher performances when including our proposed module than excluding our proposed module. The DeepConvNet with SEFE showed the highest performance of 0.72 among six different decoding models. Hence, we demonstrated the feasibility of decoding the VI dataset in the subject-independent task with robust performances by using our proposed module.