京都大学 / 0048 / 新制・課程博士 / 博士(情報学) / 甲第21213号 / 情博第666号 / 新制||情||115(附属図書館) / 京都大学大学院情報学研究科先端数理科学専攻 / (主査)講師 青柳 富誌生, 教授 西村 直志, 准教授 田口 智清, 講師 水原 啓暁 / 学位規則第4条第1項該当 / Doctor of Informatics / Kyoto University / DFAM

No description available.

No description available.

No description available.

When it comes to perception, one of the most remarkable characteristics of the brain is its omnipresent variability: Even to identical sensory stimuli, no neural response is the same. It has been hypothesized that this response variability is induced by fluctuations of the brain’s instantaneous state, yet the underlying dynamics that link such neural states with stimulus-related processes remain poorly understood. Specifically, fluctuations of excitability in sensory regions of the cortex may shape the brain’s response to external stimuli and hence the perception thereof. The current work aimed at characterizing the modulatory role and spatiotemporal organization of cortical excitability in a series of three somatosensory stimulation paradigms in humans, employing electroencephalography (EEG) to examine the interplay between pre-stimulus oscillatory state and short-latency somatosensory evoked potentials, as well as their association with the consciously accessible stimulus percept. Excitability dynamics of the primary somatosensory cortex were found to be (i) temporally structured in a special way (long-range temporal dependencies in line with the concept of criticality), (ii) linked to the behaviorally perceived stimulus intensity already through initial cortical responses, and (iii) organized with spatially confined, somatotopic patterns. Taken together, these findings suggest that fluctuations of cortical excitability reflect the maintenance of a sensitive tradeoff between robustness and flexibility of neural responses to sensory stimuli, enabling the brain to adaptively change the neural encoding of even low-level stimulus features, such as the stimulus’ intensity. Importantly, however, moment-to-moment neural response variability appears not to occur “at random”, that is, in a stochastically independent manner, but to be organized according to specific principles – both in the temporal and spatial domain.

Frontal Alpha Asymmetry (FAA) has been associated with individual differences such as various aspects of personality. However, the nature of the relationship between FAA and personality traits is not yet fully understood. The present study further investigated this relationship by exploring the correlation between resting-state FAA and the Big Five personality traits: openness, agreeableness, conscientiousness, extraversion, and neuroticism. 15 healthy participants completed resting-state EEG recordings three times and the Big Five Personality Inventory (BFI) twice. The results showed only one statistically significant correlation among the 20 correlations examined, between the F4-F3 resting-state FAA and openness scores. Besides, the direction of the relationship was the opposite of what would be expected. The small sample size of this study may have contributed to results, indicating the need for future research with larger samples. Nonetheless, the current findings add to the existing literature and suggest that the relationship between resting-state FAA and personality traits may be more complex than previously thought.

The electroencephalogram (EEG) is the primary tool used for the diagnosis of a varietyof neural pathologies such as epilepsy. Identification of a critical event, such as an epileptic seizure, is difficult because the signals are collected by transducing extremely low voltages, and as a result, are corrupted by noise. Also, EEG signals often contain artifacts due to clinical phenomena such as patient movement. These artifacts are easily confused as seizure events. Factors such as slowly evolving morphologies make accurate marking of the onset and offset of a seizure event difficult. Precise segmentation, defined as the ability to detect start and stop times within a fraction of a second, is a challenging research problem. In this dissertation, we improve seizure segmentation performance by developing deep learning technology that mimics the human interpretation process. The central thesis of this work is that separation of the seizure detection problem into a two-phase problem – epileptiform activity detection followed by seizure detection – should improve our ability to detect and localize seizure events. In the first phase, we use a sequential neural network algorithm known as a long short-term memory (LSTM) network to identify channel-specific epileptiform discharges associated with seizures. In the second phase, the feature vector is augmented with posteriors that represent the onset and offset of ictal activities. These augmented features are applied to a multichannel convolutional neural network (CNN) followed by an LSTM network. The multiphase model was evaluated on a blind evaluation set and was shown to detect 106 segment boundaries within a 2-second margin of error. Our previous best system, which delivers state-of-the-art performance on this task, correctly detected only 9 segment boundaries. Our multiphase system was also shown to be robust by performing well on two blind evaluation sets. Seizure detection performance on the TU Seizure Detection (TUSZ) Corpus development set is 41.60% sensitivity with 5.63 false alarms/24 hours (FAs/24 hrs). Performance on the corresponding evaluation set is 48.21% sensitivity with 16.54 FAs/24 hrs. Performance on a previously unseen corpus, the Duke University Seizure (DUSZ) Corpus is 46.62% sensitivity with 7.86 FAs/24 hrs. Our previous best system yields 30.83% sensitivity with 6.74 FAs/24 hrs on the TUSZ development set, 33.11% sensitivity with 19.89 FAs/24 hrs on the TUSZ evaluation set and 33.71% sensitivity with 40.40 FAs/24 hrs on DUSZ. Improving seizure detection performance through better segmentation is an important step forward in making automated seizure detection systems clinically acceptable. For a real-time system, accurate segmentation will allow clinicians detect a seizure as soon as it appears in the EEG signal. This will allow neurologists to act during the early stages of the event which, in many cases, is essential to avoid permanent damage to the brain. In a similar way, accurate offset detection will help with delivery of therapies designed to mitigate postictal (after seizure) period symptoms. This will also help reveal the severity of a seizure and consequently provide guidance for medicating a patient. / Electrical and Computer Engineering

Scalp electroencephalograms (EEGs) are used in a broad range of health care institutions to monitor and record electrical activity in the brain. EEGs are essential in diagnosis of clinical conditions such as epilepsy, seizure, coma, encephalopathy, and brain death. Manual scanning and interpretation of EEGs is time-consuming since these recordings may last hours or days. It is also an expensive process as it requires highly trained experts. Therefore, high performance automated analysis of EEGs can reduce time to diagnosis and enhance real-time applications by identifying sections of the signal that need further review.Automatic analysis of clinical EEGs is a very difficult machine learning problem due to the low fidelity of a scalp EEG signal. Commercially available automated seizure detection systems suffer from unacceptably high false alarm rates. Many signal processing methods have been developed over the years including time-frequency processing, wavelet analysis and autoregressive spectral analysis. Though there has been significant progress in machine learning technology in recent years, use of automated technology in clinical settings is limited, mainly due to unacceptably high false alarm rates. Further, state of the art machine learning algorithms that employ high dimensional models have not previously been utilized in EEG analysis because there has been a lack of large databases that accurately characterize clinical operating conditions. Deep learning approaches can be viewed as a broad family of neural network algorithms that use many layers of nonlinear processing units to learn a mapping between inputs and outputs. Deep learning-based systems have generated significant improvements in performance for sequence recognitions tasks for temporal signals such as speech and for image analysis applications that can exploit spatial correlations, and for which large amounts of training data exists. The primary goal of our proposed research is to develop deep learning-based architectures that capture spatial and temporal correlations in an EEG signal. We apply these architectures to the problem of automated seizure detection for adult EEGs. The main contribution of this work is the development of a high-performance automated EEG analysis system based on principles of machine learning and big data that approaches levels of performance required for clinical acceptance of the technology. In this work, we explore a combination of deep learning-based architectures. First, we present a hybrid architecture that integrates hidden Markov models (HMMs) for sequential decoding of EEG events with a deep learning-based postprocessing that incorporates temporal and spatial context. This system automatically processes EEG records and classifies three patterns of clinical interest in brain activity that might be useful in diagnosing brain disorders: spike and/or sharp waves, generalized periodic epileptiform discharges and periodic lateralized epileptiform discharges. It also classifies three patterns used to model the background EEG activity: eye movement, artifacts, and background. Our approach delivers a sensitivity above 90% while maintaining a specificity above 95%. Next, we replace the HMM component of the system with a deep learning architecture that exploits spatial and temporal context. We study how effectively these architectures can model context. We introduce several architectures including a novel hybrid system that integrates convolutional neural networks with recurrent neural networks to model both spatial relationships (e.g., cross-channel dependencies) and temporal dynamics (e.g., spikes). We also propose a topology-preserving architecture for spatio-temporal sequence recognition that uses raw data directly rather than low-level features. We show this model learns representations directly from raw EEGs data and does not need to use predefined features. In this study, we use the Temple University EEG (TUEG) Corpus, supplemented with data from Duke University and Emory University, to evaluate the performance of these hybrid deep structures. We demonstrate that performance of a system trained only on Temple University Seizure Corpus (TUSZ) data transfers to a blind evaluation set consisting of the Duke University Seizure Corpus (DUSZ) and the Emory University Seizure Corpus (EUSZ). This type of generalization is very important since complex high-dimensional deep learning systems tend to overtrain. We also investigate the robustness of this system to mismatched conditions (e.g., train on TUSZ, evaluate on EUSZ). We train a model on one of three available datasets and evaluate the trained model on the other two datasets. These datasets are recorded from different hospitals, using a variety of devices and electrodes, under different circumstances and annotated by different neurologists and experts. Therefore, these experiments help us to evaluate the impact of the dataset on our training process and validate our manual annotation process. Further, we introduce methods to improve generalization and robustness. We analyze performance to gain additional insight into what aspects of the signal are being modeled adequately and where the models fail. The best results for automatic seizure detection achieved in this study are 45.59% with 12.24 FA per 24 hours on TUSZ, 45.91% with 11.86 FAs on DUSZ, and 62.56% with 11.26 FAs on EUSZ. We demonstrate that the performance of the deep recurrent convolutional structure presented in this study is statistically comparable to the human performance on the same dataset. / Electrical and Computer Engineering

Objective An electroencephalographic brain–computer interface (BCI) can provide a non-muscular means of communication for people with amyotrophic lateral sclerosis (ALS) or other neuromuscular disorders. We present a novel P300-based BCI stimulus presentation – the checkerboard paradigm (CBP). CBP performance is compared to that of the standard row/column paradigm (RCP) introduced by Farwell and Donchin (1988). Methods Using an 8 × 9 matrix of alphanumeric characters and keyboard commands, 18 participants used the CBP and RCP in counter-balanced fashion. With approximately 9–12 min of calibration data, we used a stepwise linear discriminant analysis for online classification of subsequent data. Results Mean online accuracy was significantly higher for the CBP, 92%, than for the RCP, 77%. Correcting for extra selections due to errors, mean bit rate was also significantly higher for the CBP, 23 bits/min, than for the RCP, 17 bits/min. Moreover, the two paradigms produced significantly different waveforms. Initial tests with three advanced ALS participants produced similar results. Furthermore, these individuals preferred the CBP to the RCP. Conclusions These results suggest that the CBP is markedly superior to the RCP in performance and user acceptability. Significance The CBP has the potential to provide a substantially more effective BCI than the RCP. This is especially important for people with severe neuromuscular disabilities.

Attention is important in learning, Attention-deficit/hyperactivity disorder, Driving, and many other fields. Hence, intelligent tutoring systems, Attention-deficit/hyperactivity disorder diagnosis systems, and distraction detection of driver systems should be able to correctly monitor the attention levels of individuals in real time in order to estimate their attentional state. We study the feasibility of detecting distraction and concentration by monitoring participants' attention levels while they complete cognitive tasks using Electroencephalography and Eye Tracking in a virtual reality environment. Furthermore, we investigate the possibility of improving the concentration of participants using relaxation in virtual reality. We developed an indicator that estimates levels of attention with a real value using EEG data. The participant-independent indicator based on EEG data we used to assess the concentration levels of participants correctly predicts the concentration state with an accuracy (F1 = 73%). Furthermore, the participant-independent distraction model based on Eye Tracking data correctly predicted the distraction state of participants with an accuracy (F1 = 89%) in a participant-independent validation setting. / La concentration est importante dans l’apprentissage, Le trouble du déficit de l’attention avec ou sans hyperactivité, la conduite automobile et dans de nombreux autres domaines. Par conséquent, les systèmes de tutorat intelligents, les systèmes de diagnostic du trouble du déficit de l’attention avec ou sans hyperactivité et les systèmes de détection de la distraction au volant devraient être capables de surveiller correctement les niveaux d’attention des individus en temps réel afin de déduire correctement leur état attentionnel. Nous étudions la faisabilité de la détection de la distraction et de la concentration en surveillant les niveaux d’attention des participants pendant qu’ils effectuent des tâches cognitives en utilisant l’Électroencéphalographie et l’Eye Tracking dans un environnement de réalité virtuelle. En outre, nous étudions la possibilité d’améliorer la concentration des participants en utilisant la relaxation en réalité virtuelle. Nous avons mis au point un indicateur qui estime les niveaux d’attention avec une valeur réelle en utilisant les données EEG. L’indicateur indépendant du participant basé sur les données EEG que nous avons utilisé pour évaluer les niveaux de concentration des participants prédit correctement l’état de concentration avec une précision (F1 = 73%). De plus, le modèle de distraction indépendant des participants, basé sur les données d’Eye Tracking, a correctement prédit l’état de distraction des participants avec une précision (F1 = 89%) dans un cadre de validation indépendant des participants.