Decoding and interpreting cortical signals with a compact convolutional neural network
Objective: Brain-computer interfaces (BCIs) decode information from neural activity and send it to external devices. The use of Deep Learning approaches for decoding allows for automatic feature engineering within the specific decoding task. Physiologically plausible interpretation of the network parameters ensures the robustness of the learned decision rules and opens the exciting opportunity for automatic knowledge discovery. Approach: We describe a compact convolutional network-based architecture for adaptive decoding of electrocorticographic (ECoG) data into finger kinematics. We also propose a novel theoretically justified approach to interpreting the spatial and temporal weights in the architectures that combine adaptation in both space and time. The obtained spatial and frequency patterns characterizing the neuronal populations pivotal to the specific decoding task can then be interpreted by fitting appropriate spatial and dynamical models. Main results: We first tested our solution using realistic Monte-Carlo simulations. Then, when applied to the ECoG data from Berlin BCI competition IV dataset, our architecture performed comparably to the competition winners without requiring explicit feature engineering. Using the proposed approach to the network weights interpretation we could unravel the spatial and the spectral patterns of the neuronal processes underlying the successful decoding of finger kinematics from an ECoG dataset. Finally we have also applied the entire pipeline to the analysis of a 32-channel EEG motor-imagery dataset and observed physiologically plausible patterns specific to the task. Significance: We described a compact and interpretable CNN architecture derived from the basic principles and encompassing the knowledge in the field of neural electrophysiology. For the first time in the context of such multibranch architectures with factorized spatial and temporal processing we presented theoretically justified weights interpretation rules. We verified our recipes using simulations and real data and demonstrated that the proposed solution offers a good decoder and a tool for investigating motor control neural mechanisms.
Drugs and diseases play a central role in many areas of biomedical research and healthcare. Aggregating knowledge about these entities across a broader range of domains and languages is critical for information extraction (IE) applications. To facilitate text mining methods for analysis and comparison of patient’s health conditions and adverse drug reactions reported on the Internet with traditional sources such as drug labels, we present a new corpus of Russian language health reviews.
The Russian Drug Reaction Corpus (RuDReC) is a new partially annotated corpus of consumer reviews in Russian about pharmaceutical products for the detection of health-related named entities and the effectiveness of pharmaceutical products. The corpus itself consists of two parts, the raw one and the labeled one. The raw part includes 1.4 million health-related user-generated texts collected from various Internet sources, including social media. The labeled part contains 500 consumer reviews about drug therapy with drug- and disease-related information. Labels for sentences include health-related issues or their absence. The sentences with one are additionally labeled at the expression level for identification of fine-grained subtypes such as drug classes and drug forms, drug indications and drug reactions. Further, we present a baseline model for named entity recognition (NER) and multilabel sentence classification tasks on this corpus. The macro F1 score of 74.85% in the NER task was achieved by our RuDR-BERT model. For the sentence classification task, our model achieves the macro F1 score of 68.82% gaining 7.47% over the score of BERT model trained on Russian data.
This book constitutes the refereed proceedings of the 11th International Conference on Intelligent Data Processing, IDP 2016, held in Barcelona, Spain, in October 2016.
The 11 revised full papers were carefully reviewed and selected from 52 submissions. The papers of this volume are organized in topical sections on machine learning theory with applications; intelligent data processing in life and social sciences; morphological and technological approaches to image analysis.
Determining the tonality of the text is a difficult task, the solution of which essentially depends on the context, the field of study and the amount of text data. The analysis shows that the authors in their works do not jointly use the full range of possible transformations on the data and their combinations. The article explores a generalized approach, which consists in sequentially passing through the stages of intelligence analysis, obtaining a basic solution, vectorization, preprocessing, tuning hyperparameters and modeling. The experiments carried out by iterative application of these stages give a positive increase in quality for classical machine learning algorithms and a significant increase for deep learning.
Brain computer interfaces are a growing research field producing many implementations that find various uses in research and medical practice and everyday life. Despite the popularity of the implementations using non-invasive neuroimaging methods, radical improvement in the state channel bandwidth and, thus, decoding accuracy is only possible by using invasive techniques. Electrocorticography (ECoG) is a minimally invasive neuroimaging modality that provides highly informative brain activity signals and entails the use of machine learning methods to efficiently decipher the complex spatial-temporal cortical representation of motor and cognitive function. Deep learning techniques is the family of machine learning methods that allow to learn representations of data with multiple levels of abstraction. We hypothesized that the deep learning would allow to reach higher accuracy in the task of decoding movement timecourse than it is possible with traditional signal processing approaches.
article deals with the problem of isolated words recognition based on deep convolutional neural networks. The use of existing recognition systems in practice is limited by an insufficiently high degree of their reliability functioning in conditions of intense acoustic noise, such as street noise, sounds from passing vehicles, etc. Nowadays, the most accurate recognition methods are characterized by the formation of acoustic models with deep learning technologies and, in particular, convolutional neural networks. For image processing problems the possibility of adaptation of such networks to a new domain with additional finetuning on rather small training samples is well studied. In this paper we proposed to perform additional training of networks for adaptation of acoustic models on a speaker voice with use of small number of the utterances. In order to reduce the error rate, we consider an ensemble of several different speaker-dependent neural network architectures that have been trained in such a way. The final decision is made by a weighted voting rule, in which the weight of each acoustic model is determined in proportion to the accuracy estimated on the training set. The experimental results for recognition of English commands proved that such ensemble of pre-trained acoustic models can significantly improve accuracy compared to traditional pre-trained models, especially if the white Gaussian noise is added to the input signal.
Recently, deep learning methods have been increasingly applied on spoken language technologies, including signal processing, language understanding and generation, dialogue management, as well as joint optimisations of these (end-to-end learning). However, such methods still have limitations and it is not yet clear that deep learning and joint optimisation is the key to the future.
Encompassing the current deep learning trends and traditional knowledge-based methods, SLT’s 2018 main theme will be around “Spoken Language Technology in the Era of Deep Learning: Challenges and Opportunities”.
The book presents a remarkable collection of chapters covering a wide range of topics in the areas of intelligent systems and artificial intelligence, and their real-world applications. It gathers the proceedings of the Intelligent Systems Conference 2019, which attracted a total of 546 submissions from pioneering researchers, scientists, industrial engineers, and students from all around the world. These submissions underwent a double-blind peer-review process, after which 190 were selected for inclusion in these proceedings.
As intelligent systems continue to replace and sometimes outperform human intelligence in decision-making processes, they have made it possible to tackle a host of problems more effectively. This branching out of computational intelligence in several directions and use of intelligent systems in everyday applications have created the need for an international conference as a venue for reporting on the latest innovations and trends.
This book collects both theory and application based chapters on virtually all aspects of artificial intelligence; presenting state-of-the-art intelligent methods and techniques for solving real-world problems, along with a vision for future research, it represents a unique and valuable asset.
Polar mesocyclones (MCs) are small marine atmospheric vortices. The class of intense MCs, called polar lows, are accompanied by extremely strong surface winds and heat fluxes and thus largely influencing deep ocean water formation in the polar regions. Accurate detection of polar mesocyclones in high-resolution satellite data, while challenging, is a time-consuming task, when performed manually. Existing algorithms for the automatic detection of polar mesocyclones are based on the conventional analysis of patterns of cloudiness and they involve different empirically defined thresholds of geophysical variables. As a result, various detection methods typically reveal very different results when applied to a single dataset. We develop a conceptually novel approach for the detection of MCs based on the use of deep convolutional neural networks (DCNNs). As a first step, we demonstrate that DCNN model is capable of performing binary classification of 500 × 500 km patches of satellite images regarding MC patterns presence in it. The training dataset is based on the reference database of MCs manually tracked in the Southern Hemisphere from satellite mosaics. We use a subset of this database with MC diameters falling in the range of 200–400 km. This dataset is further used for testing several different DCNN setups, specifically, DCNN built “from scratch”, DCNN based on VGG16 pre-trained weights also engaging the Transfer Learning technique, and DCNN based on VGG16 with Fine Tuning technique. Each of these networks is further applied to both infrared (IR) and a combination of infrared and water vapor (IR + WV) satellite imagery. The best skills (97% in terms of the binary classification accuracy score) is achieved with the model that averages the estimates of the ensemble of different DCNNs. The algorithm can be further extended to the automatic identification and tracking numerical scheme and applied to other atmospheric phenomena that are characterized by a distinct signature in satellite imagery.
Increasing evidence suggests that neuronal communication is a defining property of functionally specialized brain networks and that it is implemented through synchronization between population activities of distinct brain areas. The detection of long-range coupling in electroencephalography (EEG) and magnetoencephalography (MEG) data using conventional metrics (such as coherence or phase-locking value) is by definition contaminated by spatial leakage. Methods such as imaginary coherence, phase-lag index or orthogonalized amplitude correlations tackle spatial leakage by ignoring zero-phase interactions. Although useful, these metrics will by construction lead to false negatives in cases where true zero-phase coupling exists in the data and will underestimate interactions with phase lags in the vicinity of zero. Yet, empirically observed neuronal synchrony in invasive recordings indicates that it is not uncommon to find zero or close-to-zero phase lag between the activity profiles of coupled neuronal assemblies. Here, we introduce a novel method that allows us to mitigate the undesired spatial leakage effects and detect zero and near zero phase interactions. To this end, we propose a projection operation that operates on sensor-space cross-spectrum and suppresses the spatial leakage contribution but retains the true zero-phase interaction component. We then solve the network estimation task as a source estimation problem defined in the product space of interacting source topographies. We show how this framework provides reliable interaction detection for all phase-lag values and we thus refer to the method as Phase Shift Invariant Imaging of Coherent Sources (PSIICOS). Realistic simulations demonstrate that PSIICOS has better detector characteristics than existing interaction metrics. Finally, we illustrate the performance of PSIICOS by applying it to real MEG dataset recorded during a standard mental rotation task. Taken together, using analytical derivations, data simulations and real brain data, this study presents a novel source-space MEG/EEG connectivity method that overcomes previous limitations and for the first time allows for the estimation of true zero-phase coupling via non-invasive electrophysiological recordings.
The series “Advances in Intelligent Systems and Computing” contains publications on theory, applications, and design methods of Intelligent Systems and Intelligent Computing. Virtually all disciplines such as engineering, natural sciences, computer and information science, ICT, economics, business, e-commerce, environment, healthcare, life science are covered. The list of topics spans all the areas of modern intelligent systems and computing such as: computational intelligence, soft computing including neural networks, fuzzy systems, evolutionary computing and the fusion of these paradigms, social intelligence, ambient intelligence, computational neuroscience, artificial life, virtual worlds and society, cognitive science and systems, Perception and Vision, DNA and immune based systems, self-organizing and adaptive systems, e-Learning and teaching, human-centered and human-centric computing, recommender systems, intelligent control, robotics and mechatronics including human-machine teaming, knowledge-based paradigms, learning paradigms, machine ethics, intelligent data analysis, knowledge management, intelligent agents, intelligent decision making and support, intelligent network security, trust management, interactive entertainment, Web intelligence and multimedia.
The publications within “Advances in Intelligent Systems and Computing” are primarily proceedings of important conferences, symposia and congresses. They cover significant recent developments in the field, both of a foundational and applicable character. An important characteristic feature of the series is the short publication time and world-wide distribution. This permits a rapid and broad dissemination of research results.
This book contains a selection of papers accepted for the presentation and discussion at the 2018 International Conference on Digital Science (DSIC’18). This Conference had the support of the Institute of Certified Specialists, Russia, AISTI (Iberian Association for Information Systems and Technologies), and Springer. It will take place at Convention Centre, Budva, Montenegro, October 19–21, 2018. DSIC’18 is an international forum for researchers and practitioners to present and discuss the most recent innovations, trends, results, experiences, and concerns in the several perspectives of Digital Science. The main idea of this Conference is that the world of science is unified and united allowing all scientists/practitioners to be able to think, analyze, and generalize their thoughts. DSIC aims efficiently to disseminate original research results in natural, social, art, and humanities sciences. An important characteristic feature of the Conference should be the short publication time and worldwide distribution. This Conference enables fast dissemination, so conference participants can publish their papers in print and electronic format, which is then made available worldwide and accessible by numerous researchers. The Scientific Committee of DSIC’18 was composed of a multidisciplinary group of 26 experts. One hundred and seven invited reviewers who are intimately concerned with Digital Science have had the responsibility for evaluating, in a “double-blind review” process, the papers received for each of the main themes proposed for the Conference: Digital Art and Humanities; Digital Economics; Digital Education; Digital Engineering; Digital Environmental Sciences; Digital Finance, Business and Banking; Digital Media; Digital Medicine, Pharma and Public Health; Digital Public Administration; Digital Technology and Applied Sciences.
DSIC’18 received 88 contributions from 16 countries around the world. The papers accepted for the presentation and discussion at the Conference are published by Springer (this book) and will be submitted for indexing by ISI, SCOPUS, among others.
Cell lines represent convenient models to elucidate specific causes of multigenetic and pluricausal diseases, to test breakthrough regenerative technologies. Most commonly used cell lines surpass diploid cells in their accessibility for delivery of large DNA molecules and genome editing, but the main obstacles for obtaining cell models with knockout-targeted protein from aneuploid cells are multiple allele copies and karyotype/phenotype heterogeneity. In the study, we report an original approach to CRISPR-/Cas9-mediated genome modification of aneuploid cell cultures to create functional cell models, achieving highly efficient targeted protein knockout and avoiding "clonal effect" (for the first time to our knowledge
Nucleic acids labeled with a fluorophore/quencher pair are widely used as probes in biomedical research and molecular diagnostics. Here we synthesized novel DNA molecular beacons double labeled with the identical dyes (R6G, ROX and Cy5) at 5′- and 3′-end and studied their photo physical properties. We demonstrated that fluorescence quenching by formation of the homo dimer exciton in such molecular beacons allows using them in homogeneous assays. Further, we developed and evaluated homo Yin-Yang DNA probes labeled with identical dyes and used them for detection of low copy HIV RNA by RT-qPCR. They demonstrated improved sensitivity (LLQ: 10 vs 30 copies mL-1) in comparison to commercially available Abbott RealTime HIV-1 kit based on VICBHQ dyes both for model mixtures (naive human plasma with added deactivated HIV-1 virus) and for preliminarily confirmed 36 clinical samples (4 vs 1 positive ones for low-copy samples).