Search

Previous research has demonstrated the potential of using pre-trained language models for decoding open vocabulary Electroencephalography (EEG) signals captured through a non-invasive Brain-Computer Interface (BCI). However, the impact of embedding EEG signals in the context of language models and the effect of subjectivity, remain unexplored, leading to uncertainty about the best approach to enhance decoding performance. Additionally, current evaluation metrics used to assess decoding effectiveness are predominantly syntactic and do not provide insights into the comprehensibility of the decoded output for human understanding. We present an end-to-end architecture for non-invasive brain recordings that brings modern representational learning approaches to neuroscience. Our proposal introduces the following innovations: 1) an end-to-end deep learning architecture for open vocabulary EEG decoding, incorporating a subject-dependent representation learning module for raw EEG encoding, a BART language model, and a GPT-4 sentence refinement module; 2) a more comprehensive sentence-level evaluation metric based on the BERTScore; 3) an ablation study that analyses the contributions of each module within our proposal, providing valuable insights for future research. We evaluate our approach on two publicly available datasets, ZuCo v1.0 and v2.0, comprising EEG recordings of 30 subjects engaged in natural reading tasks. Our model achieves a BLEU-1 score of 42.75%, a ROUGE-1-F of 33.28%, and a BERTScore-F of 53.86%, achieving an increment over the previous state-of-the-art by 1.40%, 2.59%, and 3.20%, respectively. IEEE

Graph neural networks (GNNs) are widely used for analyzing graph-structural data and solving graph-related tasks due to their powerful expressiveness. However, existing off-the-shelf GNN-based models usually consist of no more than three layers. Deeper GNNs usually suffer from severe performance degradation due to several issues including the infamous "over-smoothing" issue, which restricts the further development of GNNs. In this article, we investigate the over-smoothing issue in deep GNNs. We discover that over-smoothing not only results in indistinguishable embeddings of graph nodes, but also alters and even corrupts their semantic structures, dubbed semantic over-smoothing. Existing techniques, e.g., graph normalization, aim at handling the former concern, but neglect the importance of preserving the semantic structures in the spatial domain, which hinders the further improvement of model performance. To alleviate the concern, we propose a cluster-keeping sparse aggregation strategy to preserve the semantic structure of embeddings in deep GNNs (especially for spatial GNNs). Particularly, our strategy heuristically redistributes the extent of aggregations for all the nodes from layers, instead of aggregating them equally, so that it enables aggregate concise yet meaningful information for deep layers. Without any bells and whistles, it can be easily implemented as a plug-and-play structure of GNNs via weighted residual connections. Last, we analyze the over-smoothing issue on the GNNs with weighted residual structures and conduct experiments to demonstrate the performance comparable to the state-of-the-arts.

DNA N4-methylcytosine (4mC) site identification is a crucial field in bioinformatics, where machine learning methods have been effectively utilized. Due to the presence of noise, the existing deep learning methods for detecting 4mC have consistently low recognition rates in positive samples. With fuzzy rules and membership functions, fuzzy systems can achieve good results in processing noisy signals. In contrast to traditional fuzzy systems that lack deep feature representation and sample measurement, we introduce novel techniques to enhance generalization and feature representation. By incorporating the neural tangent kernel (NTK) and kernel learning algorithm into the fuzzy system, we propose the fuzzy neural tangent kernel (FNTK) model and the radius-based FNTK (R-FNTK) model to predict DNA 4mC sites. To achieve better generalization performance than traditional kernel functions, we first train the NTK for feature representation learning and sample measurement. Based on the membership function and NTK matrix, different fuzzy kernel matrices are constructed for each fuzzy subset of the fuzzy system. Finally, we utilize two types of iterative kernel optimization algorithms to effectively fuse multiple NTK-based fuzzy kernels and obtain the final prediction model. Rigorous testing using 6 benchmark datasets demonstrates the superiority of our approach, yielding significant improvements in the experiment's performance. © IEEE, Funding: The National Natural Science Foundation of China (NSFC 62250028, 62322215, 62172076 and U22A2038), the Excellent Young Scientists Fund in Hunan Province (Grant No. 2022JJ20077), the Zhejiang Provincial Natural Science Foundation of China (Grant No. LY23F020003) and the Municipal Government of Quzhou (2023D038).

Brain hemorrhages have become increasingly common and can be fatal if left untreated. Current methods for monitoring the progression of the disorder that rely on MRI and PET scans are inconvenient and costly for patients. This has spurred research toward portable and cost-effective techniques for predicting the current stage and malignancy of the hemorrhages. In this study, simulated S-parameter data obtained from a two-antenna system placed over the head is used in conjunction with machine learning to detect the dielectric changes in the brain caused by hemorrhage non-invasively. Several machine learning classifiers are used to analyze the data, and their performance metrics are compared to determine the optimal classifier for this case. The study revealed that Decision Tree, KNN, and Random Forest classifiers are better than SVM and MLP classifiers in terms of accuracy, precision, and recall in predicting Brain hemorrhage at the most probable locations. Contrary to conventional microwave imaging systems requiring several antennas for brain hemorrhage detection, this study demonstrates that integrating machine learning with microwave sensors enables accurate solutions with a reduced antenna count. The results present a transformative strategy for monitoring systems in clinics, where a simple, safe, and low-cost microwave antenna-based system can be intelligently integrated with machine learning to diagnose the presence of Brain hemorrhage.

Brain hemorrhages have become increasingly common and can be fatal if left untreated. Current methods for monitoring the progression of the disorder that rely on MRI and PET scans are inconvenient and costly for patients. This has spurred research toward portable and cost-effective techniques for predicting the current stage and malignancy of the hemorrhages. In this study, simulated S-parameter data obtained from a two-antenna system placed over the head is used in conjunction with machine learning to detect the dielectric changes in the brain caused by hemorrhage non-invasively. Several machine learning classifiers are used to analyze the data, and their performance metrics are compared to determine the optimal classifier for this case. The study revealed that Decision Tree, KNN, and Random Forest classifiers are better than SVM and MLP classifiers in terms of accuracy, precision, and recall in predicting Brain hemorrhage at the most probable locations. Contrary to conventional microwave imaging systems requiring several antennas for brain hemorrhage detection, this study demonstrates that integrating machine learning with microwave sensors enables accurate solutions with a reduced antenna count. The results present a transformative strategy for monitoring systems in clinics, where a simple, safe, and low-cost microwave antenna-based system can be intelligently integrated with machine learning to diagnose the presence of Brain hemorrhage.

Brain hemorrhages have become increasingly common and can be fatal if left untreated. Current methods for monitoring the progression of the disorder that rely on MRI and PET scans are inconvenient and costly for patients. This has spurred research toward portable and cost-effective techniques for predicting the current stage and malignancy of the hemorrhages. In this study, simulated S-parameter data obtained from a two-antenna system placed over the head is used in conjunction with machine learning to detect the dielectric changes in the brain caused by hemorrhage non-invasively. Several machine learning classifiers are used to analyze the data, and their performance metrics are compared to determine the optimal classifier for this case. The study revealed that Decision Tree, KNN, and Random Forest classifiers are better than SVM and MLP classifiers in terms of accuracy, precision, and recall in predicting Brain hemorrhage at the most probable locations. Contrary to conventional microwave imaging systems requiring several antennas for brain hemorrhage detection, this study demonstrates that integrating machine learning with microwave sensors enables accurate solutions with a reduced antenna count. The results present a transformative strategy for monitoring systems in clinics, where a simple, safe, and low-cost microwave antenna-based system can be intelligently integrated with machine learning to diagnose the presence of Brain hemorrhage.

Brain hemorrhages have become increasingly common and can be fatal if left untreated. Current methods for monitoring the progression of the disorder that rely on MRI and PET scans are inconvenient and costly for patients. This has spurred research toward portable and cost-effective techniques for predicting the current stage and malignancy of the hemorrhages. In this study, simulated S-parameter data obtained from a two-antenna system placed over the head is used in conjunction with machine learning to detect the dielectric changes in the brain caused by hemorrhage non-invasively. Several machine learning classifiers are used to analyze the data, and their performance metrics are compared to determine the optimal classifier for this case. The study revealed that Decision Tree, KNN, and Random Forest classifiers are better than SVM and MLP classifiers in terms of accuracy, precision, and recall in predicting Brain hemorrhage at the most probable locations. Contrary to conventional microwave imaging systems requiring several antennas for brain hemorrhage detection, this study demonstrates that integrating machine learning with microwave sensors enables accurate solutions with a reduced antenna count. The results present a transformative strategy for monitoring systems in clinics, where a simple, safe, and low-cost microwave antenna-based system can be intelligently integrated with machine learning to diagnose the presence of Brain hemorrhage.

Cognitive computing endeavors to construct models that emulate brain functions, which can be explored through electroencephalography (EEG). Developing precise and robust EEG classification models is crucial for advancing cognitive computing. Despite the high accuracy of supervised EEG classification models, they are constrained by labor-intensive annotations and poor generalization. Self-supervised models address these issues but encounter difficulties in matching the accuracy of supervised learning. Three challenges persist: 1) capturing temporal dependencies in EEG; 2) adapting loss functions to describe feature similarities in self-supervised models; and 3) addressing the prevalent issue of data imbalance in EEG. This study introduces the DreamCatcher Network (DCNet), a self-supervised EEG classification framework with a two-stage training strategy. The first stage extracts robust representations through contrastive learning, and the second stage transfers the representation encoder to a supervised EEG classification task. DCNet utilizes time-series contrastive learning to autonomously construct representations that comprehensively capture temporal correlations. A novel loss function, SelfDreamCatcherLoss, is proposed to evaluate the similarities between these representations and enhance the performance of DCNet. Additionally, two data augmentation methods are integrated to alleviate class imbalances. Extensive experiments show the superiority of DCNet over the current state-of-the-art models, achieving high accuracy on both the Sleep-EDF and HAR datasets. It holds substantial promise for revolutionizing sleep disorder detection and expediting the development of advanced healthcare systems driven by cognitive computing. © Copyright 2024 IEEE

Continuously analyzing medical time series as new classes emerge is meaningful for health monitoring and medical decision-making. Few-shot class-incremental learning (FSCIL) explores the classification of few-shot new classes without forgetting old classes. However, little of the existing research on FSCIL focuses on medical time series classification, which is more challenging to learn due to its large intra-class variability. In this paper, we propose a framework, the Meta self-Attention Prototype Incrementer (MAPIC) to address these problems. MAPIC contains three main modules: an embedding encoder for feature extraction, a prototype enhancement module for increasing inter-class variation, and a distance-based classifier for reducing intra-class variation. To mitigate catastrophic forgetting, MAPIC adopts a parameter protection strategy in which the parameters of the embedding encoder module are frozen at incremental stages after being trained in the base stage. The prototype enhancement module is proposed to enhance the expressiveness of prototypes by perceiving inter-class relations using a self-attention mechanism. We design a composite loss function containing the sample classification loss, the prototype non-overlapping loss, and the knowledge distillation loss, which work together to reduce intra-class variations and resist catastrophic forgetting. Experimental results on three different time series datasets show that MAPIC significantly outperforms state-of-the-art approaches by 27.99%, 18.4%, and 3.95%, respectively. IEEE

Affective annotation refers to the process of labeling media content based on the emotions they evoke. Since such experiences are inherently subjective and depend on individual differences, the central challenge is associating digital content with its affective, interindividual experience. Here, we present a first-of-its-kind methodology for affective annotation directly from brain signals by monitoring the affective experience of a crowd of individuals via functional near-infrared spectroscopy (fNIRS). An experiment is reported in which fNIRS was recorded from 31 participants to develop a brain-computer interface (BCI) for affective annotation. Brain signals evoked by images were used to draw predictions about the affective dimensions that characterize the stimuli. By combining annotations, the results show that monitoring crowd responses can draw accurate affective annotations, with performance improving significantly with increases in crowd size. Our methodology demonstrates a proof-of-concept to source affective annotations from a crowd of BCI users without requiring any auxiliary mental or physical interaction.

Affective annotation refers to the process of labeling media content based on the emotions they evoke. Since such experiences are inherently subjective and depend on individual differences, the central challenge is associating digital content with its affective, interindividual experience. Here, we present a first-of-its-kind methodology for affective annotation directly from brain signals by monitoring the affective experience of a crowd of individuals via functional near-infrared spectroscopy (fNIRS). An experiment is reported in which fNIRS was recorded from 31 participants to develop a brain-computer interface (BCI) for affective annotation. Brain signals evoked by images were used to draw predictions about the affective dimensions that characterize the stimuli. By combining annotations, the results show that monitoring crowd responses can draw accurate affective annotations, with performance improving significantly with increases in crowd size. Our methodology demonstrates a proof-of-concept to source affective annotations from a crowd of BCI users without requiring any auxiliary mental or physical interaction.

Brain hemorrhages have become increasingly common and can be fatal if left untreated. Current methods for monitoring the progression of the disorder that rely on MRI and PET scans are inconvenient and costly for patients. This has spurred research toward portable and cost-effective techniques for predicting the current stage and malignancy of the hemorrhages. In this study, simulated S-parameter data obtained from a two-antenna system placed over the head is used in conjunction with machine learning to detect the dielectric changes in the brain caused by hemorrhage non-invasively. Several machine learning classifiers are used to analyze the data, and their performance metrics are compared to determine the optimal classifier for this case. The study revealed that Decision Tree, KNN, and Random Forest classifiers are better than SVM and MLP classifiers in terms of accuracy, precision, and recall in predicting Brain hemorrhage at the most probable locations. Contrary to conventional microwave imaging systems requiring several antennas for brain hemorrhage detection, this study demonstrates that integrating machine learning with microwave sensors enables accurate solutions with a reduced antenna count. The results present a transformative strategy for monitoring systems in clinics, where a simple, safe, and low-cost microwave antenna-based system can be intelligently integrated with machine learning to diagnose the presence of Brain hemorrhage.

Speech feature extraction is critical for ASR systems. Such successful features as MFCC and PLP use filterbank techniques to model log-scaled speech perception but fail to model the adaptation of human speech perception by hearing experiences. Infant perception that is adapted by hearing speech without text may cause permanent brain state modifications (engrams) that serve as a physical fundamental basis for lifetime speech perception formation. This realization motivates us to propose to model such an unsupervised adaptation process, where adaptation denotes perception that is affected or changed by the history of experiences, with the Dirichlet Process Gaussian Mixture Model (DPGMM) and the DPGMM-RNN hybrid model to extract perceptual features to improve ASR. Our proposed features extend MFCC features with posteriorgrams extracted from the DPGMM algorithm or the DPGMM-RNN hybrid model. Our analysis shows that the DPGMM and DPGMM-RNN model perplexities agree with infant auditory perplexity to support that the proposed features are perceptual. Our ASR results verify the effectiveness of the proposed unsupervised features in such tasks as LVCSR on WSJ and ASR on noisy low-resource telephone conversations, compared with the supervised bottleneck features from Kaldi in ASR performance.

Electroencephalography (EEG) enables online monitoring brain activity, which can be used for neurofeedback. One of the growing applications of EEG neurofeedback is to facilitate meditation practice. Specifically, EEG neurofeedback can be used to alert participants whenever they get distracted during meditation practice based on changes in their brain activity. In this study, we develop machine learning models to detect moments of distraction (due to mind wandering or drowsiness) during meditation practice using EEG signals. We use EEG data of 24 participants while performing a breath focus meditation with experience sampling and extract twelve linear and nonlinear EEG features. Features are fed to ten supervised machine learning models to classify (i) Breath Focus Awake (BFA) vs Breath Focus Sleepy (BFS), and (ii) BFA vs Mind Wandering (MW). We observe that the linear features achieve a maximum accuracy of 86% for classifying awake (BFA) and sleepy (BFS), whereas non-linear features have more predictive ability for classifying between BFA and MW with a maximum accuracy of nearly 78%. In addition, visualization of unsupervised t-SNE lower embeddings supports the evidence of distinct clusters for each condition. Overall our results show that machine learning algorithms can successfully identify periods of distraction during meditation practice in novice meditators based on linear and non-linear features of the EEG signal. Consequently, our results have important implications for the development of mobile EEG neurofeedback protocols aimed at facilitating meditation practice., Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public., Design Aesthetics

Brain-Machine Interfaces (BMIs) assist paralyzed people to brain control (BC) the neuro-prosthesis continuously moving in space. During the BC process, the subject imagines the movement of the real limb and adapts the brain activity according to the sensory feedback. The neural adaptation in the closed-loop control results in complex and changing brain signals. Simultaneously, the decoder interprets the time-varying functional mapping between neural activity and continuous trajectory. It is crucial and challenging to accurately and adaptively track the mapping to help the subject accomplish the BC task with a stable performance. Existing Kalman Filter (KF) based decoders achieve continuous trajectory control by linearly interpreting neural firing observations into self-evolving prosthetic states. However, the linear neural-state mapping might not accurately reflect the movement intention of the subject. In this paper, we propose a novel method that allows subjects to achieve continuous brain control efficiently and stably. The proposed method incorporates a kernel reinforcement learning method into a state-observation model to decode the nonlinearly neural observation into a continuous trajectory state. The state transition function ensures the continuity of the prosthetic state. The kernel reinforcement learning allows the quick adaptation of the nonlinear neural-movement mapping during the BC process. The proposed method is tested in an online brain control reaching task for rats. Compared with KF, our method achieved more successful trials, faster response time, shorter inter-trial time, and remained stable over days. These results demonstrate that the proposed method is an efficient tool to assist subjects in brain control tasks. Author

Existing segmentation methods for brain MRI data usually leverage 3D CNNs on 3D volumes or employ 2D CNNs on 2D image slices. We discovered that while volume-based approaches well respect spatial relationships across slices, slice-based methods typically excel at capturing fine local features. Furthermore, there is a wealth of complementary information between their segmentation predictions. Inspired by this observation, we develop an Uncertainty-aware Multi-dimensional Mutual learning framework to learn different dimensional networks simultaneously, each of which provides useful soft labels as supervision to the others, thus effectively improving the generalization ability. Specifically, our framework builds upon a 2D-CNN, a 2.5D-CNN, and a 3D-CNN, while an uncertainty gating mechanism is leveraged to facilitate the selection of qualified soft labels, so as to ensure the reliability of shared information. The proposed method is a general framework and can be applied to varying backbones. The experimental results on three datasets demonstrate that our method can significantly enhance the performance of the backbone network by notable margins, achieving a Dice metric improvement of 2.8% on MeniSeg, 1.4% on IBSR, and 1.3% on BraTS2020.

Sleep-stage and apnea-hypopnea index (AHI) are the most important metrics in the diagnosis of sleep syndrome disease. In previous studies, these two tasks are usually implemented separately, which is both time- and resource-consuming. In this work, we propose a novel single electroencephalogram (EEG)-based collaborative learning network (EEG-CLNet) for simultaneous sleep staging and obstructive sleep apnea (OSA) event detection through multitask collaborative learning. The EEG-CLNet regards different tasks as a common unit to extract features from intragroups via both local parameter sharing and cross-task knowledge distillation (CTKD), rather than just sharing parameters or shortening the distance between different tasks. Our approach has been validated on two datasets with the same or better performance than other methods. The experimental results show that our method achieves a performance gain of 1%-5% compared with the baseline. Compared to previous works where two or even more models were required to perform sleep staging and OSA event detection, the EEG-CLNet could reduce the total number of model parameters and facilitate the model to mine the hidden relationships between different task semantic information. More importantly, it effectively alleviates the task bias problem in hard parameter sharing. As a consequence, this approach has notable potential to be a solution for a lightweight wearable sleep monitoring system in the future.

Identifiability of a single module in a network of transfer functions is determined by whether a particular transfer function in the network can be uniquely distinguished within a network model set, on the basis of data. Whereas previous research has focused on the situations that all network signals are either excited or measured, we develop generalized analysis results for the situation of partial measurement and partial excitation. As identifiability conditions typically require a sufficient number of external excitation signals, this article introduces a novel network model structure such that excitation from unmeasured noise signals is included, which leads to less conservative identifiability conditions than relying on measured excitation signals only. More importantly, graphical conditions are developed to verify global and generic identifiability of a single module based on the topology of the dynamic network. Depending on whether the input or the output of the module can be measured, we present four identifiability conditions which cover all possible situations in single module identification. These conditions further lead to synthesis approaches for allocating excitation signals and selecting measured signals, to warrant single module identifiability. In addition, if the identifiability conditions are satisfied for a sufficient number of external excitation signals only, indirect identification methods are developed to provide a consistent estimate of the module. All the obtained results are also extended to identifiability of multiple modules in the network., Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public., Team Bart De Schutter

This study investigates the relationship between deep learning models and the human brain, specifically focusing on the prediction of brain activity in response to static visual stimuli using functional magnetic resonance imaging (fMRI). By leveraging intermediate outputs of pre-trained convolutional neural networks (CNNs) with feature-weighted receptive fields, it becomes possible to estimate brain activity in the visual cortex. The primary objective of this research is to analyze how different CNN architectures affect the accuracy of predicting brain activity. To accomplish this, we utilize the novel fMRI Natural Scenes Dataset, which provides a large-scale data set for comprehensive analysis. Through this investigation, we aim to gain insights into the impact of CNN architectures on the prediction accuracy of brain activity in the context of visual stimuli., CSE3000 Research Project, Computer Science and Engineering

Identifiability of a single module in a network of transfer functions is determined by whether a particular transfer function in the network can be uniquely distinguished within a network model set, on the basis of data. Whereas previous research has focused on the situations that all network signals are either excited or measured, we develop generalized analysis results for the situation of partial measurement and partial excitation. As identifiability conditions typically require a sufficient number of external excitation signals, this article introduces a novel network model structure such that excitation from unmeasured noise signals is included, which leads to less conservative identifiability conditions than relying on measured excitation signals only. More importantly, graphical conditions are developed to verify global and generic identifiability of a single module based on the topology of the dynamic network. Depending on whether the input or the output of the module can be measured, we present four identifiability conditions which cover all possible situations in single module identification. These conditions further lead to synthesis approaches for allocating excitation signals and selecting measured signals, to warrant single module identifiability. In addition, if the identifiability conditions are satisfied for a sufficient number of external excitation signals only, indirect identification methods are developed to provide a consistent estimate of the module. All the obtained results are also extended to identifiability of multiple modules in the network.

Brain tumor segmentation (BTS) in magnetic resonance image (MRI) is crucial for brain tumor diagnosis, cancer management and research purposes. With the great success of the ten-year BraTS challenges as well as the advances of CNN and Transformer algorithms, a lot of outstanding BTS models have been proposed to tackle the difficulties of BTS in different technical aspects. However, existing studies hardly consider how to fuse the multi-modality images in a reasonable manner. In this paper, we leverage the clinical knowledge of how radiologists diagnose brain tumors from multiple MRI modalities and propose a clinical knowledge-driven brain tumor segmentation model, called CKD-TransBTS. Instead of directly concatenating all the modalities, we re-organize the input modalities by separating them into two groups according to the imaging principle of MRI. A dual-branch hybrid encoder with the proposed modality-correlated cross-attention block (MCCA) is designed to extract the multi-modality image features. The proposed model inherits the strengths from both Transformer and CNN with the local feature representation ability for precise lesion boundaries and long-range feature extraction for 3D volumetric images. To bridge the gap between Transformer and CNN features, we propose a Trans&CNN Feature Calibration block (TCFC) in the decoder. We compare the proposed model with six CNN-based models and six transformer-based models on the BraTS 2021 challenge dataset. Extensive experiments demonstrate that the proposed model achieves state-of-the-art brain tumor segmentation performance compared with all the competitors.

The isocitrate dehydrogenase (IDH) gene mutation is an essential biomarker for the diagnosis and prognosis of glioma. It is promising to better predict glioma genotype by integrating focal tumor image and geometric features with brain network features derived from MRI. Convolutional neural networks show reasonable performance in predicting IDH mutation, which, however, cannot learn from non-Euclidean data, e.g., geometric and network data. In this study, we propose a multi-modal learning framework using three separate encoders to extract features of focal tumor image, tumor geometrics and global brain networks. To mitigate the limited availability of diffusion MRI, we develop a self-supervised approach to generate brain networks from anatomical multi-sequence MRI. Moreover, to extract tumor-related features from the brain network, we design a hierarchical attention module for the brain network encoder. Further, we design a bi-level multi-modal contrastive loss to align the multi-modal features and tackle the domain gap at the focal tumor and global brain. Finally, we propose a weighted population graph to integrate the multi-modal features for genotype prediction. Experimental results on the testing set show that the proposed model outperforms the baseline deep learning models. The ablation experiments validate the performance of different components of the framework. The visualized interpretation corresponds to clinical knowledge with further validation. In conclusion, the proposed learning framework provides a novel approach for predicting the genotype of glioma.

Research and development of electroencephalogram (EEG) based brain-computer interfaces (BCIs) have advanced rapidly, partly due to deeper understanding of the brain and wide adoption of sophisticated machine learning approaches for decoding the EEG signals. However, recent studies have shown that machine learning algorithms are vulnerable to adversarial attacks. This paper proposes to use narrow period pulse for poisoning attack of EEG-based BCIs, which makes adversarial attacks much easier to implement. One can create dangerous backdoors in the machine learning model by injecting poisoning samples into the training set. Test samples with the backdoor key will then be classified into the target class specified by the attacker. What most distinguishes our approach from previous ones is that the backdoor key does not need to be synchronized with the EEG trials, making it very easy to implement. The effectiveness and robustness of the backdoor attack approach is demonstrated, highlighting a critical security concern for EEG-based BCIs and calling for urgent attention to address it.

Sleep staging is the process by which an overnight polysomnographic measurement is segmented into epochs of 30 seconds, each of which is annotated as belonging to one of five discrete sleep stages. The resulting scoring is graphically depicted as a hypnogram, and several overnight sleep statistics are derived, such as total sleep time and sleep onset latency. Gold standard sleep staging as performed by human technicians is time-consuming, costly, and comes with imperfect inter-scorer agreement, which also results in inter-scorer disagreement about the overnight statistics. Deep learning algorithms have shown promise in automating sleep scoring, but struggle to model inter-scorer disagreement in sleep statistics. To that end, we introduce a novel technique using conditional generative models based on Normalizing Flows that permits the modeling of the inter-rater disagreement of overnight sleep statistics, termed U-Flow. We compare U-Flow to other automatic scoring methods on a hold-out test set of 70 subjects, each scored by six independent scorers. The proposed method achieves similar sleep staging performance in terms of accuracy and Cohen's kappa on the majority-voted hypnograms. At the same time, U-Flow outperforms the other methods in terms of modeling the inter-rater disagreement of overnight sleep statistics. The consequences of inter-rater disagreement about overnight sleep statistics may be great, and the disagreement potentially carries diagnostic and scientifically relevant information about sleep structure. U-Flow is able to model this disagreement efficiently and can support further investigations into the impact inter-rater disagreement has on sleep medicine and basic sleep research.

REM sleep behavior disorder (RBD) is a parasomnia with dream enactment and presence of REM sleep without atonia (RSWA). RBD diagnosed manually via polysomnography (PSG) scoring, which is time intensive. Isolated RBD (iRBD) is also associated with a high probability of conversion to Parkinson's disease. Diagnosis of iRBD is largely based on clinical evaluation and subjective PSG ratings of REM sleep without atonia. Here we show the first application of a novel spectral vision transformer (SViT) to PSG signals for detection of RBD and compare the results to the more conventional convolutional neural network architecture. The vision-based deep learning models were applied to scalograms (30 or 300 second windows) of the PSG data (EEG, EMG and EOG) and the predictions interpreted. A total of 153 RBD (96 iRBD and 57 RBD with PD) and 190 controls were included in the study and 5-fold bagged ensemble was used. Model outputs were analyzed per-patient (averaged), with regards to sleep stage, and the SViT was interpreted using integrated gradients. Models had a similar per-epoch test F1 score. However, the vision transformer had the best per-patient performance, with an F1 score 0.87. Training the SViT on channel subsets, it achieved an F1 score of 0.93 on a combination of EEG and EOG. EMG is thought to have the highest diagnostic yield, but interpretation of our model showed that high relevance was placed on EEG and EOG, indicating these channels could be included for diagnosing RBD.

In this paper we deal with a simplified anisotropic rat head phantom development and the investigation of the influence of the anisotropic white matter on electroencephalogram source localization. The proposed phantom is based on the cubic cross cell composition combined with agar mixture to set desired electrical conductivity anisotropic ratio. For the fabrication of the phantom, the 3D printed technology is exploited. Starting from a real rat brain, we proposed a simplified brain model incorporating the actual dimensions, shape and conductivity parameters of both grey and white matter containing simultaneously relevant deep-brain electrical signal sources. Five testing dipoles were located in the areas corresponding to the active brain regions. A single dipole localization error was calculated by comparing an inverse solution with a dipole position obtained from a computer tomography image. Neglecting anisotropy had a rather weak effect on localization error of a single testing dipole in our model. The reliability map was computed and interpreted in terms of spatial similarity between distributed inverse solutions involving isotropic and anisotropic forward models. We found spatially specific error increases located close to the electrodes and in the vicinity of anisotropic compartment. Hence, areas to be most sensitive to neglecting anisotropy in our model were identified.

In this paper we deal with a simplified anisotropic rat head phantom development and the investigation of the influence of the anisotropic white matter on electroencephalogram source localization. The proposed phantom is based on the cubic cross cell composition combined with agar mixture to set desired electrical conductivity anisotropic ratio. For the fabrication of the phantom, the 3D printed technology is exploited. Starting from a real rat brain, we proposed a simplified brain model incorporating the actual dimensions, shape and conductivity parameters of both grey and white matter containing simultaneously relevant deep-brain electrical signal sources. Five testing dipoles were located in the areas corresponding to the active brain regions. A single dipole localization error was calculated by comparing an inverse solution with a dipole position obtained from a computer tomography image. Neglecting anisotropy had a rather weak effect on localization error of a single testing dipole in our model. The reliability map was computed and interpreted in terms of spatial similarity between distributed inverse solutions involving isotropic and anisotropic forward models. We found spatially specific error increases located close to the electrodes and in the vicinity of anisotropic compartment. Hence, areas to be most sensitive to neglecting anisotropy in our model were identified.

In this paper we deal with a simplified anisotropic rat head phantom development and the investigation of the influence of the anisotropic white matter on electroencephalogram source localization. The proposed phantom is based on the cubic cross cell composition combined with agar mixture to set desired electrical conductivity anisotropic ratio. For the fabrication of the phantom, the 3D printed technology is exploited. Starting from a real rat brain, we proposed a simplified brain model incorporating the actual dimensions, shape and conductivity parameters of both grey and white matter containing simultaneously relevant deep-brain electrical signal sources. Five testing dipoles were located in the areas corresponding to the active brain regions. A single dipole localization error was calculated by comparing an inverse solution with a dipole position obtained from a computer tomography image. Neglecting anisotropy had a rather weak effect on localization error of a single testing dipole in our model. The reliability map was computed and interpreted in terms of spatial similarity between distributed inverse solutions involving isotropic and anisotropic forward models. We found spatially specific error increases located close to the electrodes and in the vicinity of anisotropic compartment. Hence, areas to be most sensitive to neglecting anisotropy in our model were identified.

In this paper we deal with a simplified anisotropic rat head phantom development and the investigation of the influence of the anisotropic white matter on electroencephalogram source localization. The proposed phantom is based on the cubic cross cell composition combined with agar mixture to set desired electrical conductivity anisotropic ratio. For the fabrication of the phantom, the 3D printed technology is exploited. Starting from a real rat brain, we proposed a simplified brain model incorporating the actual dimensions, shape and conductivity parameters of both grey and white matter containing simultaneously relevant deep-brain electrical signal sources. Five testing dipoles were located in the areas corresponding to the active brain regions. A single dipole localization error was calculated by comparing an inverse solution with a dipole position obtained from a computer tomography image. Neglecting anisotropy had a rather weak effect on localization error of a single testing dipole in our model. The reliability map was computed and interpreted in terms of spatial similarity between distributed inverse solutions involving isotropic and anisotropic forward models. We found spatially specific error increases located close to the electrodes and in the vicinity of anisotropic compartment. Hence, areas to be most sensitive to neglecting anisotropy in our model were identified.

Speech feature extraction is critical for ASR systems. Such successful features as MFCC and PLP use filterbank techniques to model log-scaled speech perception but fail to model the adaptation of human speech perception by hearing experiences. Infant perception that is adapted by hearing speech without text may cause permanent brain state modifications (engrams) that serve as a physical fundamental basis for lifetime speech perception formation. This realization motivates us to propose to model such an unsupervised adaptation process, where adaptation denotes perception that is affected or changed by the history of experiences, with the Dirichlet Process Gaussian Mixture Model (DPGMM) and the DPGMM-RNN hybrid model to extract perceptual features to improve ASR. Our proposed features extend MFCC features with posteriorgrams extracted from the DPGMM algorithm or the DPGMM-RNN hybrid model. Our analysis shows that the DPGMM and DPGMM-RNN model perplexities agree with infant auditory perplexity to support that the proposed features are perceptual. Our ASR results verify the effectiveness of the proposed unsupervised features in such tasks as LVCSR on WSJ and ASR on noisy low-resource telephone conversations, compared with the supervised bottleneck features from Kaldi in ASR performance.

In this paper we deal with a simplified anisotropic rat head phantom development and the investigation of the influence of the anisotropic white matter on electroencephalogram source localization. The proposed phantom is based on the cubic cross cell composition combined with agar mixture to set desired electrical conductivity anisotropic ratio. For the fabrication of the phantom, the 3D printed technology is exploited. Starting from a real rat brain, we proposed a simplified brain model incorporating the actual dimensions, shape and conductivity parameters of both grey and white matter containing simultaneously relevant deep-brain electrical signal sources. Five testing dipoles were located in the areas corresponding to the active brain regions. A single dipole localization error was calculated by comparing an inverse solution with a dipole position obtained from a computer tomography image. Neglecting anisotropy had a rather weak effect on localization error of a single testing dipole in our model. The reliability map was computed and interpreted in terms of spatial similarity between distributed inverse solutions involving isotropic and anisotropic forward models. We found spatially specific error increases located close to the electrodes and in the vicinity of anisotropic compartment. Hence, areas to be most sensitive to neglecting anisotropy in our model were identified.

Speech feature extraction is critical for ASR systems. Such successful features as MFCC and PLP use filterbank techniques to model log-scaled speech perception but fail to model the adaptation of human speech perception by hearing experiences. Infant perception that is adapted by hearing speech without text may cause permanent brain state modifications (engrams) that serve as a physical fundamental basis for lifetime speech perception formation. This realization motivates us to propose to model such an unsupervised adaptation process, where adaptation denotes perception that is affected or changed by the history of experiences, with the Dirichlet Process Gaussian Mixture Model (DPGMM) and the DPGMM-RNN hybrid model to extract perceptual features to improve ASR. Our proposed features extend MFCC features with posteriorgrams extracted from the DPGMM algorithm or the DPGMM-RNN hybrid model. Our analysis shows that the DPGMM and DPGMM-RNN model perplexities agree with infant auditory perplexity to support that the proposed features are perceptual. Our ASR results verify the effectiveness of the proposed unsupervised features in such tasks as LVCSR on WSJ and ASR on noisy low-resource telephone conversations, compared with the supervised bottleneck features from Kaldi in ASR performance., journal article

Estimating a sequence of dynamic undirected graphical models, in which adjacent graphs share similar structures, is of paramount importance in various social, financial, biological, and engineering systems, since the evolution of such networks can be utilized for example to spot trends, detect anomalies, predict vulnerability, and evaluate the impact of interventions. Existing methods for learning dynamic graphical models require the tuning parameters that control the graph sparsity and the temporal smoothness to be selected via brute-force grid search. Furthermore, these methods are computationally burdensome with time complexity O(NP3)O(NP3) for PP variables and NN time points. As a remedy, we propose a low-complexity tuning-free Bayesian approach, named BASS. Specifically, we impose temporally dependent spike and slab priors on the graphs such that they are sparse and varying smoothly across time. An efficient variational inference algorithm based on natural gradients is then derived to learn the graph structures from the data in an automatic manner. Owing to the pseudo-likelihood and the mean-field approximation, the time complexity of BASS is only O(NP2)O(NP2). To cope with the local maxima problem of variational inference, we resort to simulated annealing and propose a method based on bootstrapping of the observations to generate the annealing noise. We provide numerical evidence that BASS outperforms existing methods on synthetic data in terms of structure estimation, while being more efficient especially when the dimension PP becomes high. We further apply the approach to the stock return data of 78 banks from 2005 to 2013 and find that the number of edges in the financial network as a function of time contains three peaks, in coincidence with the 2008 global financial crisis and the two subsequent European debt crisis. On the other hand, by identifying the frequency-domain resemblance to the time-varying graphical models, we show that BASS can be extende, Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public., Signal Processing Systems

Rapid eye movement (REM) sleep behavior disorder (RBD) is parasomnia and a prodromal manifestation of Parkinson's disease. The current diagnostic method relies on manual scoring of polysomnograms (PSGs), a procedure that is time and effort intensive, subject to interscorer variability, and requires high level of expertise. Here, we present an automatic and interpretable diagnostic tool for RBD that analyzes PSGs using end-to-end deep neural networks. We optimized hierarchical attention networks in a 5-fold cross validation directly to classify RBD from PSG data recorded in 143 participants with RBD and 147 age-and sex-matched controls. An ensemble model using logistic regression was implemented to fuse decisions from networks trained in various signal combinations. We interpreted the networks using gradient SHAP that attribute relevance of input signals to model decisions. The ensemble model achieved a sensitivity of 91.4 % and a specificity of 86.3 %. Interpretation showed that electroencephalography (EEG) and leg electromyography (EMG) exhibited most patterns with high relevance. This study validates a robust diagnostic tool for RBD and proposes an interpretable and fully automatic framework for end-to-end modeling of other sleep disorders from PSG data. Clinical relevance- This study presents a novel diagnostic tool for RBD that considers neurophysiologic biomarkers in multiple modalities.

It is of great significance to apply deep learning for the early diagnosis of Alzheimer's disease (AD). In this work, a novel tensorizing GAN with high-order pooling is proposed to assess mild cognitive impairment (MCI) and AD. By tensorizing a three-player cooperative game-based framework, the proposed model can benefit from the structural information of the brain. By incorporating the high-order pooling scheme into the classifier, the proposed model can make full use of the second-order statistics of holistic magnetic resonance imaging (MRI). To the best of our knowledge, the proposed Tensor-train, High-order pooling and Semisupervised learning-based GAN (THS-GAN) is the first work to deal with classification on MR images for AD diagnosis. Extensive experimental results on Alzheimer's disease neuroimaging initiative (ADNI) data set are reported to demonstrate that the proposed THS-GAN achieves superior performance compared with existing methods, and to show that both tensor-train and high-order pooling can enhance classification performance. The visualization of generated samples also shows that the proposed model can generate plausible samples for semisupervised learning purpose.

It is of great significance to apply deep learning for the early diagnosis of Alzheimer's disease (AD). In this work, a novel tensorizing GAN with high-order pooling is proposed to assess mild cognitive impairment (MCI) and AD. By tensorizing a three-player cooperative game-based framework, the proposed model can benefit from the structural information of the brain. By incorporating the high-order pooling scheme into the classifier, the proposed model can make full use of the second-order statistics of holistic magnetic resonance imaging (MRI). To the best of our knowledge, the proposed Tensor-train, High-order pooling and Semisupervised learning-based GAN (THS-GAN) is the first work to deal with classification on MR images for AD diagnosis. Extensive experimental results on Alzheimer's disease neuroimaging initiative (ADNI) data set are reported to demonstrate that the proposed THS-GAN achieves superior performance compared with existing methods, and to show that both tensor-train and high-order pooling can enhance classification performance. The visualization of generated samples also shows that the proposed model can generate plausible samples for semisupervised learning purpose. IEEE

It is of great significance to apply deep learning for the early diagnosis of Alzheimer's disease (AD). In this work, a novel tensorizing GAN with high-order pooling is proposed to assess mild cognitive impairment (MCI) and AD. By tensorizing a three-player cooperative game-based framework, the proposed model can benefit from the structural information of the brain. By incorporating the high-order pooling scheme into the classifier, the proposed model can make full use of the second-order statistics of holistic magnetic resonance imaging (MRI). To the best of our knowledge, the proposed Tensor-train, High-order pooling and Semisupervised learning-based GAN (THS-GAN) is the first work to deal with classification on MR images for AD diagnosis. Extensive experimental results on Alzheimer's disease neuroimaging initiative (ADNI) data set are reported to demonstrate that the proposed THS-GAN achieves superior performance compared with existing methods, and to show that both tensor-train and high-order pooling can enhance classification performance. The visualization of generated samples also shows that the proposed model can generate plausible samples for semisupervised learning purpose. IEEE

Sleep is a natural phenomenon controlled by the central nervous system. The sleep-wake pattern, which functions as an essential indicator of neurophysiological organization in the neonatal period, has profound meaning in the prediction of cognitive diseases and brain maturity. In recent years, unobtrusive sleep monitoring and automatic sleep staging have been intensively studied for adults, but much less for neonates. This work aims to investigate a novel video-based unobtrusive method for neonatal sleep-wake classification by analyzing the behavioral changes in the neonatal facial region. A hybrid model is proposed to monitor the sleep-wake patterns of human neonates. The model combines two algorithms: deep convolutional neural network (DCNN) and support vector machine (SVM), where DCNN works as a trainable feature extractor and SVM as a classifier. Data was collected from nineteen Chinese neonates at the Children's Hospital of Fudan University, Shanghai, China. The classification results are compared with the gold standard of video-electroencephalography scored by pediatric neurologists. Validations indicate that the proposed hybrid DCNN-SVM model achieved reliable performances in classifying neonatal sleep and wake states in RGB video frames (with the face region detected), with an accuracy of 93.8 ± 2.2% and an F1-score 0.93 ± 0.3.

Understanding adolescent decision-making is significant for informing basic models of neurodevelopment as well as for the domains of public health and criminal justice. System-based theories posit that adolescent decision-making is guided by activity amongst reward and control processes. While successful at explaining behavior, system-based theories have received inconsistent support at the neural level, perhaps because of methodological limitations. Here, we used two complementary approaches to overcome said limitations and rigorously evaluate system-based models. Using decision-level modeling of fMRI data from a risk-taking task in a sample of 2000+ decisions across 51 human adolescents (25 females, mean age = 15.00 years), we find support for system-based theories of decision-making. Neural activity in lateral prefrontal cortex and a multivariate pattern of cognitive control both predicted a reduced likelihood of risk-taking, whereas increased activity in the nucleus accumbens predicted a greater likelihood of risk-taking. Interactions between decision-level brain activity and age were not observed. These results garner support for system-based accounts of adolescent decision-making behavior.SIGNIFICANCE STATEMENT:Adolescent decision-making behavior is of great import for basic science, and carries equally consequential implications for public health and criminal justice. While dominant psychological theories seeking to explain adolescent decision-making have found empirical support, their neuroscientific implementations have received inconsistent support. This may be partly due to statistical approaches employed by prior neuroimaging studies of system-based theories. We used brain modeling-an approach that predicts behavior from brain activity-of univariate and multivariate neural activity metrics to better understand how neural components of psychological systems guide decision behavior in adolescents. We found broad support for system-based theories such that

Objective: Most cardiac arrest patients who are successfully resuscitated are initially comatose due to hypoxic-ischemic brain injury. Quantitative electroencephalography (EEG) provides valuable prognostic information. However, prior approaches largely rely on snapshots of the EEG, without taking advantage of temporal information. Methods: We present a recurrent deep neural network with the goal of capturing temporal dynamics from longitudinal EEG data to predict long-term neurological outcomes. We utilized a large international dataset of continuous EEG recordings from 1,038 cardiac arrest patients from seven hospitals in Europe and the US. Poor outcome was defined as a Cerebral Performance Category (CPC) score of 3-5, and good outcome as CPC score 0-2 at 3 to 6-months after cardiac arrest. Model performance is evaluated using 5-fold cross validation. Results: The proposed approach provides predictions which improve over time, beginning from an area under the receiver operating characteristic curve (AUC-ROC) of 0.78 (95% CI: 0.72-0.81) at 12 hours, and reaching 0.88 (95% CI: 0.85-0.91) by 66 h after cardiac arrest. At 66 h, (sensitivity, specificity) points of interest on the ROC curve for predicting poor outcomes were (32,99)%, (55,95)%, and (62,90)%, (99,23)%, (95,47)%, and (90,62)%; whereas for predicting good outcome, the corresponding operating points were (17,99)%, (47,95)%, (62,90)%, (99,19)%, (95,48)%, (70,90)%. Moreover, the model provides predicted probabilities that closely match the observed frequencies of good and poor outcomes (calibration error 0.04). Conclusions and Significance: These findings suggest that accounting for EEG trend information can substantially improve prediction of neurologic outcomes for patients with coma following cardiac arrest., SCOPUS: ar.j, info:eu-repo/semantics/published

Sleep is a natural phenomenon controlled by the central nervous system. The sleep-wake pattern, which functions as an essential indicator of neurophysiological organization in the neonatal period, has profound meaning in the prediction of cognitive diseases and brain maturity. In recent years, unobtrusive sleep monitoring and automatic sleep staging have been intensively studied for adults, but much less for neonates. This work aims to investigate a novel video-based unobtrusive method for neonatal sleep-wake classification by analyzing the behavioral changes in the neonatal facial region. A hybrid model is proposed to monitor the sleep-wake patterns of human neonates. The model combines two algorithms: deep convolutional neural network (DCNN) and support vector machine (SVM), where DCNN works as a trainable feature extractor and SVM as a classifier. Data was collected from nineteen Chinese neonates at the Children's Hospital of Fudan University, Shanghai, China. The classification results are compared with the gold standard of video-electroencephalography scored by pediatric neurologists. Validations indicate that the proposed hybrid DCNN-SVM model achieved reliable performances in classifying neonatal sleep and wake states in RGB video frames (with the face region detected), with an accuracy of 93.8 ± 2.2% and an F1-score 0.93 ± 0.3.

This paper investigates the possibility of using deliberated geometries for rotating targets which may improve the detectability and classification by means of its micro-Doppler signature. Beyond the signature provided by the rotation rate of the target, every particular geometry leads to specific and unique modulation waveforms that increase the information related to this target. Different star-shaped wheels with particular geometries have been designed demonstrating different reflection characteristics and modulation waveforms. The signature of these rotating star-shaped wheels has been measured in an anechoic chamber using a general purpose 24 GHz FMCW radar platform and a detection algorithm based on the derivative of the cross-correlation is proposed which validates the proposed approach.

A novel modeling approach capturing the multiple peak phenomenon in oral levodopa administration is proposed. Multiple peaks in the blood plasma concentration of the drug are attributed to the effects caused by gastric emptying. The developed model describes the instances of interrupted gastric emptying by an impulsive feedback of the dopamine concentration in the brain acting on the pyloric sphincter. A combination of the continuous levodopa clearing dynamics and the impulsive feedback results in a hybrid model, whose solutions are positive and bounded. The stability properties of the model are studied by means of a Poincaré map describing the propagation of the continuous model states through the firings of the impulsive feedback. Model feasibility is illustrated on data sets obtained in clinical experiments.

The Bayesian Confidence Propagation Neural Network (BCPNN) has been applied in higher level of cognitive intelligence (e.g. working memory, associative memory). However, in the spike-based version of this learning rule the pre-, postsynaptic and coincident activity is traced in three low-passfiltering stages, the calculation processes of weight update are very computationally intensive. In this paper, a hardware architecture of the updating process for lazy update mode is proposed for updating 8 local synaptic state variables. The parallelism by decomposing the calculation steps of formulas based on the inherent data dependencies is optimized. The FPGA-based hardware accelerator of BCPNN is designed and implemented. The experimental results show the updating process on FPGA can be accomplished within 110 ns with a clock frequency of 200 MHz, the updating speed is greatly enhanced compared with the CPU test. The trade-off between performance, accuracy and resources on dedicated hardware is evaluated, and the impact of the module reuse on resource consumption and computing performance is evaluated., QC 20211215Part of proceeding: ISBN 978-1-7281-9226-0

The efficacy of deep brain stimulation (DBS) depends on electrode placement accuracy, which can be jeopardized by brain shift due to burr hole and dura opening during surgery. Brain shift violates assumed rigid alignment between preoperative image and intraoperative anatomy, negatively impacting therapy.ObjectiveThis study presents a deformation-atlas biomechanical model-based approach to address shift.MethodsSix patients, who underwent interventional magnetic resonance (iMR) image-guided DBS burr hole surgery, were studied. A patient-specific model was employed under varying surgical conditions, generating a collection of possible intraoperative shift estimations or a 'deformation atlas.' An inverse problem was driven by sparse measurements derived from iMR to determine an optimal fit of solutions of the atlas. This fit was then used to obtain a volumetric deformation field, which was utilized to update preoperative MR and estimate shift at surgical target region localized on iMR. Model performance was examined by quantitatively comparing intraoperative subsurface measurements to their model-predicted counterparts, and qualitatively comparing iMR, preoperative MR, and model updated MR. A nonrigid image registration was introduced as a comparator.ResultsModel-based approach reduced general parenchyma shift from 8.2 ± 2.2 to 2.7 ± 1.1 mm (∼66.8% correction), and produced updated MR with better agreement to iMR than that of preoperative MR. The average model estimated shift at target region was 1.2 mm.ConclusionsThis study demonstrates the feasibility of a model-based shift correction strategy in DBS surgery with only sparse data.SignificanceThe developed strategy has the potential to complement and/or enhance current clinical approaches in addressing shift.

Electromagnetic brain imaging is the reconstruction of brain activity from non-invasive recordings of the magnetic fields and electric potentials. An enduring challenge in this imaging modality is estimating the number, location, and time course of sources, especially for the reconstruction of distributed brain sources with complex spatial extent. Here, we introduce a novel robust empirical Bayesian algorithm that enables better reconstruction of distributed brain source activity with two key ideas: kernel smoothing and hyperparameter tiling. Since the proposed algorithm builds upon many of the performance features of the sparse source reconstruction algorithm - Champagne and we refer to this algorithm as Smooth Champagne. Smooth Champagne is robust to the effects of high levels of noise, interference, and highly correlated brain source activity. Simulations demonstrate excellent performance of Smooth Champagne when compared to benchmark algorithms in accurately determining the spatial extent of distributed source activity. Smooth Champagne also accurately reconstructs real MEG and EEG data.

Brain function is hallmarked by its adaptivity and robustness, arising from underlying neural activity that admits well-structured representations in the temporal, spatial, or spectral domains. While neuroimaging techniques such as Electroencephalography (EEG) and magnetoencephalography (MEG) can record rapid neural dynamics at high temporal resolutions, they face several signal processing challenges that hinder their full utilization in capturing these characteristics of neural activity. The objective of this dissertation is to devise statistical modeling and estimation methodologies that account for the dynamic and structured representations of neural activity and to demonstrate their utility in application to experimentally-recorded data. The first part of this dissertation concerns spectral analysis of neural data. In order to capture the non-stationarities involved in neural oscillations, we integrate multitaper spectral analysis and state-space modeling in a Bayesian estimation setting. We also present a multitaper spectral analysis method tailored for spike trains that captures the non-linearities involved in neuronal spiking. We apply our proposed algorithms to both EEG and spike recordings, which reveal significant gains in spectral resolution and noise reduction. In the second part, we investigate cortical encoding of speech as manifested in MEG responses. These responses are often modeled via a linear filter, referred to as the temporal response function (TRF). While the TRFs estimated from the sensor-level MEG data have been widely studied, their cortical origins are not fully understood. We define the new notion of Neuro-Current Response Functions (NCRFs) for simultaneously determining the TRFs and their cortical distribution. We develop an efficient algorithm for NCRF estimation and apply it to MEG data, which provides new insights into the cortical dynamics underlying speech processing. Finally, in the third part, we consider the inference of Granger

This paper introduces a simple approach combining deep learning and histogram contour processing for automatic detection of various types of artifact contaminating the raw electroencephalogram (EEG). The proposed method considers both spatial and temporal information of raw EEG, without additional need for reference signals like ECG or EOG. The proposed method was evaluated with data including 785 EEG sequences contaminated by artifacts and 785 artifact-free EEG sequences collected from 15 intensive care patients. The obtained results showed an overall accuracy of 0.98, representing high reliability of proposed technique in detecting different types of artifacts and being comparable or outperforming the approaches proposed earlier in the literature.

We show that two widely accepted model reduction techniques, balanced truncation (BT) and balanced singular perturbation approximation (BSPA), can be derived as limiting approximations of a carefully constructed parameterization of linear time invariant systems by employing the model boundary approximation method (MBAM) [1]. We also show that MBAM provides a novel way to interpolate between BT and BSPA, by exploring the set of approximations on the boundary of the "model manifold," which is associated with the specific choice of model parameterization and initial condition and is embedded in a sample space of measured outputs, between the elements that correspond to the two model reduction techniques. This paper suggests similar types of approximations may be obtainable in topologically similar places (i.e., on certain boundaries) on the associated model manifold of nonlinear systems if analogous parameterizations can be achieved, therefore extending these widely accepted model reduction techniques to nonlinear systems.(1), QC 20191210