Search

Abstract Recent years have seen vast progress in the development of machine learned force fields (MLFFs) based on ab-initio reference calculations. Despite achieving low test errors, the reliability of MLFFs in molecular dynamics (MD) simulations is facing growing scrutiny due to concerns about instability over extended simulation timescales. Our findings suggest a potential connection between robustness to cumulative inaccuracies and the use of equivariant representations in MLFFs, but the computational cost associated with these representations can limit this advantage in practice. To address this, we propose a transformer architecture called SO3krates that combines sparse equivariant representations (Euclidean variables) with a self-attention mechanism that separates invariant and equivariant information, eliminating the need for expensive tensor products. SO3krates achieves a unique combination of accuracy, stability, and speed that enables insightful analysis of quantum properties of matter on extended time and system size scales. To showcase this capability, we generate stable MD trajectories for flexible peptides and supra-molecular structures with hundreds of atoms. Furthermore, we investigate the PES topology for medium-sized chainlike molecules (e.g., small peptides) by exploring thousands of minima. Remarkably, SO3krates demonstrates the ability to strike a balance between the conflicting demands of stability and the emergence of new minimum-energy conformations beyond the training data, which is crucial for realistic exploration tasks in the field of biochemistry.

We focus on detecting anomalies in images where the data distribution is supported by a lower-dimensional embedded manifold. Approaches based on autoencoders have aimed to control their capacity either by reducing the size of the bottleneck layer or by imposing sparsity constraints on their activations. However, none of these techniques explicitly penalize the reconstruction of anomalous regions, often resulting in poor detection. We tackle this problem by adapting a self-supervised learning regime that essentially implements a denoising autoencoder with structured non-i.i.d. noise. Informally, our objective is to regularize the model to produce locally consistent reconstructions while replacing irregularities by acting as a filter that removes anomalous patterns. Formally, we show that the resulting model resembles a nonlinear orthogonal projection of partially corrupted images onto the submanifold of uncorrupted examples. Furthermore, we identify the orthogonal projection as an optimal solution for a specific regularized autoencoder related to contractive and denoising variants. In addition, orthogonal projection provides a conservation effect by largely preserving the original content of its arguments. Together, these properties facilitate an accurate detection and localization of anomalous regions by means of the reconstruction error. We support our theoretical analysis by achieving state-of-the-art results (image/pixel-level AUROC of 99.8/99.2%) on the MVTec AD dataset—a challenging benchmark for anomaly detection in the manufacturing domain.

In recent years, increasingly complex machine learning methods have become state-of-the-art in modelling wind turbine power curves based on operational data. While these methods often exhibit superior performance on test sets, they face criticism due to a perceived lack of transparency and concerns about their robustness in dynamic, non-stationary environments encountered by wind turbines. In this work, we address these issues and present a framework that leverages explainable artificial intelligence methods to gain systematic insights into data-driven power curve models. At its core, we propose a metric to quantify how well a learned model strategy aligns with the underlying physical principles of the problem. This novel tool enables model validation beyond the conventional error metrics in an automated manner. We demonstrate, for instance, its capacity as an indicator for model generalization even when limited data is available. Moreover, it facilitates understanding how decisions made during the machine learning development process, such as data selection, pre-processing, or training parameters, affect learned strategies. As a result, we obtain physically more reasonable models, a prerequisite not only for robustness but also for meaningful insights into turbine operation by domain experts. The latter, we illustrate in the context of wind turbine performance monitoring. In summary, the framework aims to guide researchers and practitioners alike toward a more informed selection and utilization of data-driven wind turbine power curve models.

Sinonasal tumour diagnosis can be complicated by the heterogeneity of disease and classification systems. Here, the authors use machine learning to classify sinonasal undifferentiated carcinomas into 4 molecular classe with differences in differentiation state and clinical outcome.

Abstract Histological sections of the lymphatic system are usually the basis of static (2D) morphological investigations. Here, we performed a dynamic (4D) analysis of human reactive lymphoid tissue using confocal fluorescent laser microscopy in combination with machine learning. Based on tracks for T-cells (CD3), B-cells (CD20), follicular T-helper cells (PD1) and optical flow of follicular dendritic cells (CD35), we put forward the first quantitative analysis of movement-related and morphological parameters within human lymphoid tissue. We identified correlations of follicular dendritic cell movement and the behavior of lymphocytes in the microenvironment. In addition, we investigated the value of movement and/or morphological parameters for a precise definition of cell types (CD clusters). CD-clusters could be determined based on movement and/or morphology. Differentiating between CD3- and CD20 positive cells is most challenging and long term-movement characteristics are indispensable. We propose morphological and movement-related prototypes of cell entities applying machine learning models. Finally, we define beyond CD clusters new subgroups within lymphocyte entities based on long term movement characteristics. In conclusion, we showed that the combination of 4D imaging and machine learning is able to define characteristics of lymphocytes not visible in 2D histology.

Single-pulse electrical stimulation in the nervous system, often called cortico-cortical evoked potential (CCEP) measurement, is an important technique to understand how brain regions interact with one another. Voltages are measured from implanted electrodes in one brain area while stimulating another with brief current impulses separated by several seconds. Historically, researchers have tried to understand the significance of evoked voltage polyphasic deflections by visual inspection, but no general-purpose tool has emerged to understand their shapes or describe them mathematically. We describe and illustrate a new technique to parameterize brain stimulation data, where voltage response traces are projected into one another using a semi-normalized dot product. The length of timepoints from stimulation included in the dot product is varied to obtain a temporal profile of structural significance, and the peak of the profile uniquely identifies the duration of the response. Using linear kernel PCA, a canonical response shape is obtained over this duration, and then single-trial traces are parameterized as a projection of this canonical shape with a residual term. Such parameterization allows for dissimilar trace shapes from different brain areas to be directly compared by quantifying cross-projection magnitudes, response duration, canonical shape projection amplitudes, signal-to-noise ratios, explained variance, and statistical significance. Artifactual trials are automatically identified by outliers in sub-distributions of cross-projection magnitude, and rejected. This technique, which we call "Canonical Response Parameterization" (CRP) dramatically simplifies the study of CCEP shapes, and may also be applied in a wide range of other settings involving event-triggered data.

Most machine-learning force fields dismiss long-range interactions. Here the authors demonstrate the BIGDML approach for building materials’ potential energy surfaces that enables a broad range of materials simulations within accuracies better than 1 meV/atom using just 10–200 structures for training.

Abstract Understanding the pathological properties of dysregulated protein networks in individual patients’ tumors is the basis for precision therapy. Functional experiments are commonly used, but cover only parts of the oncogenic signaling networks, whereas methods that reconstruct networks from omics data usually only predict average network features across tumors. Here, we show that the explainable AI method layer-wise relevance propagation (LRP) can infer protein interaction networks for individual patients from proteomic profiling data. LRP reconstructs average and individual interaction networks with an AUC of 0.99 and 0.93, respectively, and outperforms state-of-the-art network prediction methods for individual tumors. Using data from The Cancer Proteome Atlas, we identify known and potentially novel oncogenic network features, among which some are cancer-type specific and show only minor variation among patients, while others are present across certain tumor types but differ among individual patients. Our approach may therefore support predictive diagnostics in precision oncology by inferring “patient-level” oncogenic mechanisms.

Molecular relaxation, finding the equilibrium state of a non-equilibrium structure, is an essential component of computational chemistry to understand reactivity. Classical force field (FF) methods often rely on insufficient local energy minimization, while neural network FF models require large labeled datasets encompassing both equilibrium and non-equilibrium structures. As a remedy, we propose MoreRed, molecular relaxation by reverse diffusion, a conceptually novel and purely statistical approach where non-equilibrium structures are treated as noisy instances of their corresponding equilibrium states. To enable the denoising of arbitrarily noisy inputs via a generative diffusion model, we further introduce a novel diffusion time step predictor. Notably, MoreRed learns a simpler pseudo potential energy surface (PES) instead of the complex physical PES. It is trained on a significantly smaller, and thus computationally cheaper, dataset consisting of solely unlabeled equilibrium structures, avoiding the computation of non-equilibrium structures altogether. We compare MoreRed to classical FFs, equivariant neural network FFs trained on a large dataset of equilibrium and non-equilibrium data, as well as a semi-empirical tight-binding model. To assess this quantitatively, we evaluate the root-mean-square deviation between the found equilibrium structures and the reference equilibrium structures as well as their energies.

The targeted discovery of molecules with specific structural and chemical properties is an open challenge in computational chemistry. Here, the authors propose a conditional generative neural network for the inverse design of 3d molecular structures.

Current machine-learned force fields typically ignore electronic degrees of freedom. SpookyNet is a deep neural network that explicitly treats electronic degrees of freedom, closing an important remaining gap for models in quantum chemistry.

Brain-age (BA) estimates based on deep learning are increasingly used as neuroimaging biomarker for brain health; however, the underlying neural features have remained unclear. We combined ensembles of convolutional neural networks with Layer-wise Relevance Propagation (LRP) to detect which brain features contribute to BA. Trained on magnetic resonance imaging (MRI) data of a population-based study (n = 2637, 18–82 years), our models estimated age accurately based on single and multiple modalities, regionally restricted and whole-brain images (mean absolute errors 3.37–3.86 years). We find that BA estimates capture ageing at both small and large-scale changes, revealing gross enlargements of ventricles and subarachnoid spaces, as well as white matter lesions, and atrophies that appear throughout the brain. Divergence from expected ageing reflected cardiovascular risk factors and accelerated ageing was more pronounced in the frontal lobe. Applying LRP, our study demonstrates how superior deep learning models detect brain-ageing in healthy and at-risk individuals throughout adulthood.

Considering the worst-case scenario, the junction-tree algorithm remains the most general solution for exact MAP inference with polynomial run-time guarantees. Unfortunately, its main tractability assumption requires the treewidth of a corresponding MRF to be bounded, strongly limiting the range of admissible applications. In fact, many practical problems in the area of structured prediction require modeling global dependencies by either directly introducing global factors or enforcing global constraints on the prediction variables. However, this always results in a fully-connected graph, making exact inferences by means of this algorithm intractable. Previous works focusing on the problem of loss-augmented inference have demonstrated how efficient inference can be performed on models with specific global factors representing non-decomposable loss functions within the training regime of SSVMs. Making the observation that the same fundamental idea can be applied to solve a broader class of computational problems, in this paper, we adjust the framework for an efficient exact inference to allow much finer interactions between the energy of the core model and the sufficient statistics of the global terms. As a result, we greatly increase the range of admissible applications and strongly improve upon the theoretical guarantees of computational efficiency. We illustrate the applicability of our method in several use cases, including one that is not covered by the previous problem formulation. Furthermore, we propose a new graph transformation technique via node cloning, which ensures a polynomial run-time for solving our target problem. In particular, the overall computational complexity of our constrained message-passing algorithm depends only on form-independent quantities such as the treewidth of a corresponding graph (without global connections) and image size of the sufficient statistics of the global terms.

Machine learning is an established and frequently used technique in industry and academia, but a standard process model to improve success and efficiency of machine learning applications is still missing. Project organizations and machine learning practitioners face manifold challenges and risks when developing machine learning applications and have a need for guidance to meet business expectations. This paper therefore proposes a process model for the development of machine learning applications, covering six phases from defining the scope to maintaining the deployed machine learning application. Business and data understanding are executed simultaneously in the first phase, as both have considerable impact on the feasibility of the project. The next phases are comprised of data preparation, modeling, evaluation, and deployment. Special focus is applied to the last phase, as a model running in changing real-time environments requires close monitoring and maintenance to reduce the risk of performance degradation over time. With each task of the process, this work proposes quality assurance methodology that is suitable to address challenges in machine learning development that are identified in the form of risks. The methodology is drawn from practical experience and scientific literature, and has proven to be general and stable. The process model expands on CRISP-DM, a data mining process model that enjoys strong industry support, but fails to address machine learning specific tasks. The presented work proposes an industry- and application-neutral process model tailored for machine learning applications with a focus on technical tasks for quality assurance.

The brain-computer interface (BCI) has been investigated as a form of communication tool between the brain and external devices. BCIs have been extended beyond communication and control over the years. The 2020 international BCI competition aimed to provide high-quality neuroscientific data for open access that could be used to evaluate the current degree of technical advances in BCI. Although there are a variety of remaining challenges for future BCI advances, we discuss some of more recent application directions: (i) few-shot EEG learning, (ii) micro-sleep detection (iii) imagined speech decoding, (iv) cross-session classification, and (v) EEG(+ear-EEG) detection in an ambulatory environment. Not only did scientists from the BCI field compete, but scholars with a broad variety of backgrounds and nationalities participated in the competition to address these challenges. Each dataset was prepared and separated into three data that were released to the competitors in the form of training and validation sets followed by a test set. Remarkable BCI advances were identified through the 2020 competition and indicated some trends of interest to BCI researchers.

Cross-frequency synchronization (CFS) has been proposed as a mechanism for integrating spatially and spectrally distributed information in the brain. However, investigating CFS in Magneto- and Electroencephalography (MEG/EEG) is hampered by the presence of spurious neuronal interactions due to the non-sinusoidal waveshape of brain oscillations. Such waveshape gives rise to the presence of oscillatory harmonics mimicking genuine neuronal oscillations. Until recently, however, there has been no methodology for removing these harmonics from neuronal data. In order to address this long-standing challenge, we introduce a novel method (called HARMOnic miNImization - Harmoni) that removes the signal components which can be harmonics of a non-sinusoidal signal. Harmoni’s working principle is based on the presence of CFS between harmonic components and the fundamental component of a non-sinusoidal signal. We extensively tested Harmoni in realistic EEG simulations. The simulated couplings between the source signals represented genuine and spurious CFS and within-frequency phase synchronization. Using diverse evaluation criteria, including ROC analyses, we showed that the within- and cross-frequency spurious interactions are suppressed significantly, while the genuine activities are not affected. Additionally, we applied Harmoni to real resting-state EEG data revealing intricate remote connectivity patterns which are usually masked by the spurious connections. Given the ubiquity of non-sinusoidal neuronal oscillations in electrophysiological recordings, Harmoni is expected to facilitate novel insights into genuine neuronal interactions in various research fields, and can also serve as a steppingstone towards the development of further signal processing methods aiming at refining within- and cross-frequency synchronization in electrophysiological recordings.

The inclusion of nuclear quantum effects (NQE) in atomistic simulations of chemical systems is of key importance. Here the authors use machine learned force fields trained on coupled cluster reference data to show the dynamical strengthening of covalent and non-covalent molecular interactions induced by NQE.

High-level ab initio quantum chemical methods carry a high computational burden, thus limiting their applicability. Here, the authors employ machine learning to generate coupled-cluster energies and forces at chemical accuracy for geometry optimization and molecular dynamics from DFT densities.

Abstract Digital contact tracing approaches based on Bluetooth low energy (BLE) have the potential to efficiently contain and delay outbreaks of infectious diseases such as the ongoing SARS-CoV-2 pandemic. In this work we propose a machine learning based approach to reliably detect subjects that have spent enough time in close proximity to be at risk of being infected. Our study is an important proof of concept that will aid the battery of epidemiological policies aiming to slow down the rapid spread of COVID-19.

There is an increasing number of medical use cases where classification algorithms based on deep neural networks reach performance levels that are competitive with human medical experts. To alleviate the challenges of small dataset sizes, these systems often rely on pretraining. In this work, we aim to assess the broader implications of these approaches in order to better understand what type of pretraining works reliably (with respect to performance, robustness, learned representation etc.) in practice and what type of pretraining dataset is best suited to achieve good performance in small target dataset size scenarios. Considering diabetic retinopathy grading as an exemplary use case, we compare the impact of different training procedures including recently established self-supervised pretraining methods based on contrastive learning. To this end, we investigate different aspects such as quantitative performance, statistics of the learned feature representations, interpretability and robustness to image distortions. Our results indicate that models initialized from ImageNet pretraining report a significant increase in performance, generalization and robustness to image distortions. In particular, self-supervised models show further benefits to supervised models. Self-supervised models with initialization from ImageNet pretraining not only report higher performance, they also reduce overfitting to large lesions along with improvements in taking into account minute lesions indicative of the progression of the disease. Understanding the effects of pretraining in a broader sense that goes beyond simple performance comparisons is of crucial importance for the broader medical imaging community beyond the use case considered in this work.

Methods for electro- or magnetoencephalography (EEG/MEG) based brain source imaging (BSI) using sparse Bayesian learning (SBL) have been demonstrated to achieve excellent performance in situations with low numbers of distinct active sources, such as event-related designs. This paper extends the theory and practice of SBL in three important ways. First, we reformulate three existing SBL algorithms under the majorization-minimization (MM) framework. This unification perspective not only provides a useful theoretical framework for comparing different algorithms in terms of their convergence behavior, but also provides a principled recipe for constructing novel algorithms with specific properties by designing appropriate bounds of the Bayesian marginal likelihood function. Second, building on the MM principle, we propose a novel method called LowSNR-BSI that achieves favorable source reconstruction performance in low signal-to-noise-ratio (SNR) settings. Third, precise knowledge of the noise level is a crucial requirement for accurate source reconstruction. Here we present a novel principled technique to accurately learn the noise variance from the data either jointly within the source reconstruction procedure or using one of two proposed cross-validation strategies. Empirically, we could show that the monotonous convergence behavior predicted from MM theory is confirmed in numerical experiments. Using simulations, we further demonstrate the advantage of LowSNR-BSI over conventional SBL in low-SNR regimes, and the advantage of learned noise levels over estimates derived from baseline data. To demonstrate the usefulness of our novel approach, we show neurophysiologically plausible source reconstructions on averaged auditory evoked potential data.

Brain networks can be explored by delivering brief pulses of electrical current in one area while measuring voltage responses in other areas. We propose a convergent paradigm to study brain dynamics, focusing on a single brain site to observe the average effect of stimulating each of many other brain sites. Viewed in this manner, visually-apparent motifs in the temporal response shape emerge from adjacent stimulation sites. This work constructs and illustrates a data-driven approach to determine characteristic spatiotemporal structure in these response shapes, summarized by a set of unique "basis profile curves" (BPCs). Each BPC may be mapped back to underlying anatomy in a natural way, quantifying projection strength from each stimulation site using simple metrics. Our technique is demonstrated for an array of implanted brain surface electrodes in a human patient. This framework enables straightforward interpretation of single-pulse brain stimulation data, and can be applied generically to explore the diverse milieu of interactions that comprise the connectome.

Nonlinear machine learning methods have good predictive ability but the lack of transparency of the algorithms can limit their use. Here the authors investigate how these methods approach learning in order to assess the dependability of their decision making.

Brain-Computer Interfaces (BCIs) are systems that allow users to control devices using brain activity alone. However, the ability of participants to command BCIs varies from subject to subject. About 20% of potential users of sensorimotor BCIs do not gain reliable control of the system. The inefficiency to decode user's intentions requires the identification of neurophysiological factors determining “good” and “poor” BCI performers. One of the important neurophysiological aspects in BCI research is that the neuronal oscillations, used to control these systems, show a rich repertoire of spatial sensorimotor interactions. Considering this, we hypothesized that neuronal connectivity in sensorimotor areas would define BCI performance. Analyses for this study were performed on a large dataset of 80 inexperienced participants. They took part in a calibration and an online feedback session recorded on the same day. Undirected functional connectivity was computed over sensorimotor areas by means of the imaginary part of coherency. The results show that post- as well as pre-stimulus connectivity in the calibration recording is significantly correlated to online feedback performance in μ and feedback frequency bands. Importantly, the significance of the correlation between connectivity and BCI feedback accuracy was not due to the signal-to-noise ratio of the oscillations in the corresponding post and pre-stimulus intervals. Thus, this study demonstrates that BCI performance is not only dependent on the amplitude of sensorimotor oscillations as shown previously, but that it also relates to sensorimotor connectivity measured during the preceding training session. The presence of such connectivity between motor and somatosensory systems is likely to facilitate motor imagery, which in turn is associated with the generation of a more pronounced modulation of sensorimotor oscillations (manifested in ERD/ERS) required for the adequate BCI performance. We also discuss strategies for the up-regulation of such connectivity in order to enhance BCI performance.

Cross-frequency coupling (CFC) between neuronal oscillations reflects an integration of spatially and spectrally distributed information in the brain. Here, we propose a novel framework for detecting such interactions in Magneto- and Electroencephalography (MEG/EEG), which we refer to as Nonlinear Interaction Decomposition (NID). In contrast to all previous methods for separation of cross-frequency (CF) sources in the brain, we propose that the extraction of nonlinearly interacting oscillations can be based on the statistical properties of their linear mixtures. The main idea of NID is that nonlinearly coupled brain oscillations can be mixed in such a way that the resulting linear mixture has a non-Gaussian distribution. We evaluate this argument analytically for amplitude-modulated narrow-band oscillations which are either phase-phase or amplitude-amplitude CF coupled. We validated NID extensively with simulated EEG obtained with realistic head modelling. The method extracted nonlinearly interacting components reliably even at SNRs as small as −15 dB. Additionally, we applied NID to the resting-state EEG of 81 subjects to characterize CF phase-phase coupling between alpha and beta oscillations. The extracted sources were located in temporal, parietal and frontal areas, demonstrating the existence of diverse local and distant nonlinear interactions in resting-state EEG data. All codes are available publicly via GitHub.

For the robust estimation of evoked brain activity from functional Near-Infrared Spectroscopy (fNIRS) signals, it is crucial to reduce nuisance signals from systemic physiology and motion. The current best practice incorporates short-separation (SS) fNIRS measurements as regressors in a General Linear Model (GLM). However, several challenging signal characteristics such as non-instantaneous and non-constant coupling are not yet addressed by this approach and additional auxiliary signals are not optimally exploited. We have recently introduced a new methodological framework for the unsupervised multivariate analysis of fNIRS signals using Blind Source Separation (BSS) methods. Building onto the framework, in this manuscript we show how to incorporate the advantages of regularized temporally embedded Canonical Correlation Analysis (tCCA) into the supervised GLM. This approach allows flexible integration of any number of auxiliary modalities and signals. We provide guidance for the selection of optimal parameters and auxiliary signals for the proposed GLM extension. Its performance in the recovery of evoked HRFs is then evaluated using both simulated ground truth data and real experimental data and compared with the GLM with short-separation regression. Our results show that the GLM with tCCA significantly improves upon the current best practice, yielding significantly better results across all applied metrics: Correlation (HbO max. +45%), Root Mean Squared Error (HbO max. −55%), F-Score (HbO up to 3.25-fold) and p-value as well as power spectral density of the noise floor. The proposed method can be incorporated into the GLM in an easily applicable way that flexibly combines any available auxiliary signals into optimal nuisance regressors. This work has potential significance both for conventional neuroscientific fNIRS experiments as well as for emerging applications of fNIRS in everyday environments, medicine and BCI, where high Contrast to Noise Ratio is of importance for single trial analysis.

Abstract Background Comprehensive mutational profiling data now available on all major cancers have led to proposals of novel molecular tumor classifications that modify or replace the established organ- and tissue-based tumor typing. The rationale behind such molecular reclassifications is that genetic alterations underlying cancer pathology predict response to therapy and may therefore offer a more precise view on cancer than histology. The use of individual actionable mutations to select cancers for treatment across histotypes is already being tested in the so-called basket trials with variable success rates. Here, we present a computational approach that facilitates the systematic analysis of the histological context dependency of mutational effects by integrating genomic and proteomic tumor profiles across cancers. Methods To determine effects of oncogenic mutations on protein profiles, we used the energy distance, which compares the Euclidean distances of protein profiles in tumors with an oncogenic mutation (inner distance) to that in tumors without the mutation (outer distance) and performed Monte Carlo simulations for the significance analysis. Finally, the proteins were ranked by their contribution to profile differences to identify proteins characteristic of oncogenic mutation effects across cancers. Results We apply our approach to four current proposals of molecular tumor classifications and major therapeutically relevant actionable genes. All 12 actionable genes evaluated show effects on the protein level in the corresponding tumor type and showed additional mutation-related protein profiles in 21 tumor types. Moreover, our analysis identifies consistent cross-cancer effects for 4 genes (FGFR1, ERRB2, IDH1, KRAS/NRAS) in 14 tumor types. We further use cell line drug response data to validate our findings. Conclusions This computational approach can be used to identify mutational signatures that have protein-level effects and can therefore contribute to preclinical in silico tests of the efficacy of molecular classifications as well as the druggability of individual mutations. It thus supports the identification of novel targeted therapies effective across cancers and guides efficient basket trial designs.

Simultaneous accurate and efficient prediction of molecular properties relies on combined quantum mechanics and machine learning approaches. Here the authors develop a flexible machine-learning force-field with high-level accuracy for molecular dynamics simulations.

Machine learning allows electronic structure calculations to access larger system sizes and, in dynamical simulations, longer time scales. Here, the authors perform such a simulation using a machine-learned density functional that avoids direct solution of the Kohn-Sham equations.

The service quality and system dependability of real-time communication networks strongly depends on the analysis of monitored data, to identify concrete problems and their causes. Many of these can be described by either their structural or temporal properties, or a combination of both. As current research is short of approaches sufficiently addressing both properties simultaneously, we propose a new feature space specifically suited for this task, which we analyze for its theoretical properties and its practical relevance. We evaluate its classification performance when used on real-world data sets of structural-temporal mobile communication data, and compare it to the performance achieved of feature representations used in related work. For this purpose we propose a system which allows the automatic detection and prediction of classes of pre-defined sequence behavior, greatly reducing costs caused by the otherwise required manual analysis. With our proposed feature spaces this system achieves a precision of more than 93% at recall values of 100%, with an up to 6.7% higher effective recall than otherwise similarly performing alternatives, notably outperforming alternative deep learning, kernel learning and ensemble learning approaches of related work. Furthermore the supported system calibration allows separating reliable from unreliable predictions more effectively, which is highly relevant for any practical application.

The application of deep learning (DL) models to neuroimaging data poses several challenges, due to the high dimensionality, low sample size, and complex temporo-spatial dependency structure of these data. Even further, DL models often act as black boxes, impeding insight into the association of cognitive state and brain activity. To approach these challenges, we introduce the DeepLight framework, which utilizes long short-term memory (LSTM) based DL models to analyze whole-brain functional Magnetic Resonance Imaging (fMRI) data. To decode a cognitive state (e.g., seeing the image of a house), DeepLight separates an fMRI volume into a sequence of axial brain slices, which is then sequentially processed by an LSTM. To maintain interpretability, DeepLight adapts the layer-wise relevance propagation (LRP) technique. Thereby, decomposing its decoding decision into the contributions of the single input voxels to this decision. Importantly, the decomposition is performed on the level of single fMRI volumes, enabling DeepLight to study the associations between cognitive state and brain activity on several levels of data granularity, from the level of the group down to the level of single time points. To demonstrate the versatility of DeepLight, we apply it to a large fMRI dataset of the Human Connectome Project. We show that DeepLight outperforms conventional approaches of uni- and multivariate fMRI analysis in decoding the cognitive states and in identifying the physiologically appropriate brain regions associated with these states. We further demonstrate DeepLight's ability to study the fine-grained temporo-spatial variability of brain activity over sequences of single fMRI samples.

More than ever, technical inventions are the symbol of our society's advance. Patents guarantee their creators protection against infringement. For an invention being patentable, its novelty and inventiveness have to be assessed. Therefore, a search for published work that describes similar inventions to a given patent application needs to be performed. Currently, this so-called search for prior art is executed with semi-automatically composed keyword queries, which is not only time consuming, but also prone to errors. In particular, errors may systematically arise by the fact that different keywords for the same technical concepts may exist across disciplines. In this paper, a novel approach is proposed, where the full text of a given patent application is compared to existing patents using machine learning and natural language processing techniques to automatically detect inventions that are similar to the one described in the submitted document. Various state-of-the-art approaches for feature extraction and document comparison are evaluated. In addition to that, the quality of the current search process is assessed based on ratings of a domain expert. The evaluation results show that our automated approach, besides accelerating the search process, also improves the search results for prior art with respect to their quality.

Brain-Computer Interfaces (BCIs) are inefficient for a non-negligible part of the population, estimated around 25%. To understand this phenomenon in Sensorimotor Rhythm (SMR) based BCIs, data from a large-scale screening study conducted on 80 novice participants with the Berlin BCI system and its standard machine-learning approach were investigated. Each participant performed one BCI session with resting state Encephalography, Motor Observation, Motor Execution and Motor Imagery recordings and 128 electrodes. A significant portion of the participants (40%) could not achieve BCI control (feedback performance > 70%). Based on the performance of the calibration and feedback runs, BCI users were stratified in three groups. Analyses directed to detect and elucidate the differences in the SMR activity of these groups were performed. Statistics on reactive frequencies, task prevalence and classification results are reported. Based on their SMR activity, also a systematic list of potential reasons leading to performance drops and thus hints for possible improvements of BCI experimental design are given. The categorization of BCI users has several advantages, allowing researchers 1) to select subjects for further analyses as well as for testing new BCI paradigms or algorithms, 2) to adopt a better subject-dependent training strategy and 3) easier comparisons between different studies.

Brain-computer interfaces (BCIs) have been studied extensively in order to establish a non-muscular communication channel mainly for patients with impaired motor functions. However, many limitations remain for BCIs in clinical use. In this study, we propose a hybrid BCI that is based on only frontal brain areas and can be operated in an eyes-closed state for end users with impaired motor and declining visual functions. In our experiment, electroencephalography (EEG) and near-infrared spectroscopy (NIRS) were simultaneously measured while 12 participants performed mental arithmetic (MA) and remained relaxed (baseline state: BL). To evaluate the feasibility of the hybrid BCI, we classified MA- from BL-related brain activation. We then compared classification accuracies using two unimodal BCIs (EEG and NIRS) and the hybrid BCI in an offline mode. The classification accuracy of the hybrid BCI (83.9 ± 10.3%) was shown to be significantly higher than those of unimodal EEG-based (77.3 ± 15.9%) and NIRS-based BCI (75.9 ± 6.3%). The analytical results confirmed performance improvement with the hybrid BCI, particularly for only frontal brain areas. Our study shows that an eyes-closed hybrid BCI approach based on frontal areas could be applied to neurodegenerative patients who lost their motor functions, including oculomotor functions.

Chance favors the prepared mind, said Louis Pasteur. Sometimes, significant breakthroughs occur when we creatively integrate new information, leading to a creative insight or an Aha! moment, while at other times when we fail to use a clue, we remain stuck in our habitual thinking patterns. In this study, we hypothesized that the brain's transient oscillatory states would characterize its receptivity or preparedness for such insights. We conducted a real-time brain-state-dependent cognitive stimulation experiment during insightful problem-solving. We showed that participants were more successful in utilizing clues and experienced more Aha responses when these clues were presented at the spontaneously up-regulated state of right temporal alpha oscillation, as opposed to the down-regulated state. Furthermore, we observed an inverse correlation between the coupling of alpha oscillation phase and gamma oscillation power and the frequency of insight. These results shed light on the neural mechanism underpinning the brain's receptivity to integrate upcoming semantic information, emphasizing the pivotal role of dynamical brain oscillations in the Aha! experience.

The robust analysis of neural signals is a challenging problem. Here, we contribute a convolutional neural network (CNN) for the robust classification of a steady-state visual evoked potentials (SSVEPs) paradigm. We measure electroencephalogram (EEG)-based SSVEPs for a brain-controlled exoskeleton under ambulatory conditions in which numerous artifacts may deteriorate decoding. The proposed CNN is shown to achieve reliable performance under these challenging conditions. To validate the proposed method, we have acquired an SSVEP dataset under two conditions: 1) a static environment, in a standing position while fixated into a lower-limb exoskeleton and 2) an ambulatory environment, walking along a test course wearing the exoskeleton (here, artifacts are most challenging). The proposed CNN is compared to a standard neural network and other state-of-the-art methods for SSVEP decoding (i.e., a canonical correlation analysis (CCA)-based classifier, a multivariate synchronization index (MSI), a CCA combined with k-nearest neighbors (CCA-KNN) classifier) in an offline analysis. We found highly encouraging SSVEP decoding results for the CNN architecture, surpassing those of other methods with classification rates of 99.28% and 94.03% in the static and ambulatory conditions, respectively. A subsequent analysis inspects the representation found by the CNN at each layer and can thus contribute to a better understanding of the CNN's robust, accurate decoding abilities.

High prediction accuracies are not the only objective to consider when solving problems using machine learning. Instead, particular scientific applications require some explanation of the learned prediction function. For computational biology, positional oligomer importance matrices (POIMs) have been successfully applied to explain the decision of support vector machines (SVMs) using weighted-degree (WD) kernels. To extract relevant biological motifs from POIMs, the motifPOIM method has been devised and showed promising results on real-world data. Our contribution in this paper is twofold: as an extension to POIMs, we propose gPOIM, a general measure of feature importance for arbitrary learning machines and feature sets (including, but not limited to, SVMs and CNNs) and devise a sampling strategy for efficient computation. As a second contribution, we derive a convex formulation of motifPOIMs that leads to more reliable motif extraction from gPOIMs. Empirical evaluations confirm the usefulness of our approach on artificially generated data as well as on real-world datasets.

There are some practical factors, such as arm position change and donning/doffing, which prevent robust myoelectric control. The objective of this study is to precisely characterize the impacts of the two representative factors on myoelectric controllability in practical control situations, thereby providing useful references that can be potentially used to find better solutions for clinically reliable myoelectric control. To this end, a real-time target acquisition task was performed by fourteen subjects including one individual with congenital upper-limb deficiency, where the impacts of arm position change, donning/doffing and a combination of both factors on control performance was systematically evaluated. The changes in online performance were examined with seven different performance metrics to comprehensively evaluate various aspects of myoelectric controllability. As a result, arm position change significantly affects offline prediction accuracy, but not online control performance due to real-time feedback, thereby showing no significant correlation between offline and online performance. Donning/doffing was still problematic in online control conditions. It was further observed that no benefit was attained when using a control model trained with multiple position data in terms of arm position change, and the degree of electrode shift caused by donning/doffing was not severely associated with the degree of performance loss under practical conditions (around 1 cm electrode shift). Since this study is the first to concurrently investigate the impacts of arm position change and donning/doffing in practical myoelectric control situations, all findings of this study provide new insights into robust myoelectric control with respect to arm position change and donning/doffing.

OBJECTIVE:Using traditional approaches, a brain-computer interface (BCI) requires the collection of calibration data for new subjects prior to online use. Calibration time can be reduced or eliminated e.g., by subject-to-subject transfer of a pre-trained classifier or unsupervised adaptive classification methods which learn from scratch and adapt over time. While such heuristics work well in practice, none of them can provide theoretical guarantees. Our objective is to modify an event-related potential (ERP) paradigm to work in unison with the machine learning decoder, and thus to achieve a reliable unsupervised calibrationless decoding with a guarantee to recover the true class means. METHOD:We introduce learning from label proportions (LLP) to the BCI community as a new unsupervised, and easy-to-implement classification approach for ERP-based BCIs. The LLP estimates the mean target and non-target responses based on known proportions of these two classes in different groups of the data. We present a visual ERP speller to meet the requirements of LLP. For evaluation, we ran simulations on artificially created data sets and conducted an online BCI study with 13 subjects performing a copy-spelling task. RESULTS:Theoretical considerations show that LLP is guaranteed to minimize the loss function similar to a corresponding supervised classifier. LLP performed well in simulations and in the online application, where 84.5% of characters were spelled correctly on average without prior calibration. SIGNIFICANCE:The continuously adapting LLP classifier is the first unsupervised decoder for ERP BCIs guaranteed to find the optimal decoder. This makes it an ideal solution to avoid tedious calibration sessions. Additionally, LLP works on complementary principles compared to existing unsupervised methods, opening the door for their further enhancement when combined with LLP.

On sedation motivated by the clinical needs for safety and reliability, recent studies have attempted to identify brain-specific signatures for tracking patient transition into and out of consciousness, but the differences in neurophysiological effects between 1) the sedative types and 2) the presence/absence of surgical stimulations still remain unclear. Here we used multimodal electroencephalography-functional near-infrared spectroscopy (EEG-fNIRS) measurements to observe electrical and hemodynamic responses during sedation simultaneously. Forty healthy volunteers were instructed to push the button to administer sedatives in response to auditory stimuli every 9-11 s. To generally illustrate brain activity at repetitive transition points at the loss of consciousness (LOC) and the recovery of consciousness (ROC), patient-controlled sedation was performed using two different sedatives (midazolam (MDZ) and propofol (PPF)) under two surgical conditions. Once consciousness was lost via sedatives, we observed gradually increasing EEG power at lower frequencies (15 Hz), as well as spatially increased EEG powers in the delta and lower alpha bands, and particularly also in the upper alpha rhythm, at the frontal and parieto-occipital areas over time. During ROC from unconsciousness, these spatio-temporal changes were reversed. Interestingly, the level of consciousness was switched on/off at significantly higher effect-site concentrations of sedatives in the brain according to the use of surgical stimuli, but the spatio-temporal EEG patterns were similar, regardless of the sedative used. We also observed sudden phase shifts in fronto-parietal connectivity at the LOC and the ROC as critical points. fNIRS measurement also revealed mild hemodynamic fluctuations. Compared with general anesthesia, our results provide insights into critical hallmarks of sedative-induced (un)consciousness, which have similar spatio-temporal EEG-fNIRS patterns regardless of the stage and the sedative used.

Text documents can be described by a number of abstract concepts such as semantic category, writing style, or sentiment. Machine learning (ML) models have been trained to automatically map documents to these abstract concepts, allowing to annotate very large text collections, more than could be processed by a human in a lifetime. Besides predicting the text's category very accurately, it is also highly desirable to understand how and why the categorization process takes place. In this paper, we demonstrate that such understanding can be achieved by tracing the classification decision back to individual words using layer-wise relevance propagation (LRP), a recently developed technique for explaining predictions of complex non-linear classifiers. We train two word-based ML models, a convolutional neural network (CNN) and a bag-of-words SVM classifier, on a topic categorization task and adapt the LRP method to decompose the predictions of these models onto words. Resulting scores indicate how much individual words contribute to the overall classification decision. This enables one to distill relevant information from text documents without an explicit semantic information extraction step. We further use the word-wise relevance scores for generating novel vector-based document representations which capture semantic information. Based on these document vectors, we introduce a measure of model explanatory power and show that, although the SVM and CNN models perform similarly in terms of classification accuracy, the latter exhibits a higher level of explainability which makes it more comprehensible for humans and potentially more useful for other applications.

Electroencephalography (EEG) and near-infrared spectroscopy (NIRS) are non-invasive neuroimaging methods that record the electrical and metabolic activity of the brain, respectively. Hybrid EEG-NIRS brain-computer interfaces (hBCIs) that use complementary EEG and NIRS information to enhance BCI performance have recently emerged to overcome the limitations of existing unimodal BCIs, such as vulnerability to motion artifacts for EEG-BCI or low temporal resolution for NIRS-BCI. However, with respect to NIRS-BCI, in order to fully induce a task-related brain activation, a relatively long trial length (≥10 s) is selected owing to the inherent hemodynamic delay that lowers the information transfer rate (ITR; bits/min). To alleviate the ITR degradation, we propose a more practical hBCI operated by intuitive mental tasks, such as mental arithmetic (MA) and word chain (WC) tasks, performed within a short trial length (5 s). In addition, the suitability of the WC as a BCI task was assessed, which has so far rarely been used in the BCI field. In this experiment, EEG and NIRS data were simultaneously recorded while participants performed MA and WC tasks without preliminary training and remained relaxed (baseline; BL). Each task was performed for 5 s, which was a shorter time than previous hBCI studies. Subsequently, a classification was performed to discriminate MA-related or WC-related brain activations from BL-related activations. By using hBCI in the offline/pseudo-online analyses, average classification accuracies of 90.0 ± 7.1/85.5 ± 8.1% and 85.8 ± 8.6/79.5 ± 13.4% for MA vs. BL and WC vs. BL, respectively, were achieved. These were significantly higher than those of the unimodal EEG- or NIRS-BCI in most cases. Given the short trial length and improved classification accuracy, the average ITRs were improved by more than 96.6% for MA vs. BL and 87.1% for WC vs. BL, respectively, compared to those reported in previous studies. The suitability of implementing a more practical hBCI based on intuitive mental tasks without preliminary training and with a shorter trial length was validated when compared to previous studies.

Identifying discriminative motifs underlying the functionality and evolution of organisms is a major challenge in computational biology. Machine learning approaches such as support vector machines (SVMs) achieve state-of-the-art performances in genomic discrimination tasks, but--due to its black-box character--motifs underlying its decision function are largely unknown. As a remedy, positional oligomer importance matrices (POIMs) allow us to visualize the significance of position-specific subsequences. Although being a major step towards the explanation of trained SVM models, they suffer from the fact that their size grows exponentially in the length of the motif, which renders their manual inspection feasible only for comparably small motif sizes, typically k ≤ 5. In this work, we extend the work on positional oligomer importance matrices, by presenting a new machine-learning methodology, entitled motifPOIM, to extract the truly relevant motifs--regardless of their length and complexity--underlying the predictions of a trained SVM model. Our framework thereby considers the motifs as free parameters in a probabilistic model, a task which can be phrased as a non-convex optimization problem. The exponential dependence of the POIM size on the oligomer length poses a major numerical challenge, which we address by an efficient optimization framework that allows us to find possibly overlapping motifs consisting of up to hundreds of nucleotides. We demonstrate the efficacy of our approach on a synthetic data set as well as a real-world human splice site data set.

Understanding and interpreting classification decisions of automated image classification systems is of high value in many applications, as it allows to verify the reasoning of the system and provides additional information to the human expert. Although machine learning methods are solving very successfully a plethora of tasks, they have in most cases the disadvantage of acting as a black box, not providing any information about what made them arrive at a particular decision. This work proposes a general solution to the problem of understanding classification decisions by pixel-wise decomposition of nonlinear classifiers. We introduce a methodology that allows to visualize the contributions of single pixels to predictions for kernel-based classifiers over Bag of Words features and for multilayered neural networks. These pixel contributions can be visualized as heatmaps and are provided to a human expert who can intuitively not only verify the validity of the classification decision, but also focus further analysis on regions of potential interest. We evaluate our method for classifiers trained on PASCAL VOC 2009 images, synthetic image data containing geometric shapes, the MNIST handwritten digits data set and for the pre-trained ImageNet model available as part of the Caffe open source package.

Classifying multivariate electromyography (EMG) data is an important problem in prosthesis control as well as in neurophysiological studies and diagnosis. With modern high-density EMG sensor technology, it is possible to capture the rich spectrospatial structure of the myoelectric activity. We hypothesize that multi-way machine learning methods can efficiently utilize this structure in classification as well as reveal interesting patterns in it. To this end, we investigate the suitability of existing three-way classification methods to EMG-based hand movement classification in spectrospatial domain, as well as extend these methods by sparsification and regularization. We propose to use Fourier-domain independent component analysis as preprocessing to improve classification and interpretability of the results. In high-density EMG experiments on hand movements across 10 subjects, three-way classification yielded higher average performance compared with state-of-the art classification based on temporal features, suggesting that the three-way analysis approach can efficiently utilize detailed spectrospatial information of high-density EMG. Phase and amplitude patterns of features selected by the classifier in finger-movement data were found to be consistent with known physiology. Thus, our approach can accurately resolve hand and finger movements on the basis of detailed spectrospatial information, and at the same time allows for physiological interpretation of the results.

Brain-Computer Interfaces (BCIs) strive to decode brain signals into control commands for severely handicapped people with no means of muscular control. These potential users of noninvasive BCIs display a large range of physical and mental conditions. Prior studies have shown the general applicability of BCI with patients, with the conflict of either using many training sessions or studying only moderately restricted patients. We present a BCI system designed to establish external control for severely motor-impaired patients within a very short time. Within only six experimental sessions, three out of four patients were able to gain significant control over the BCI, which was based on motor imagery or attempted execution. For the most affected patient, we found evidence that the BCI could outperform the best assistive technology (AT) of the patient in terms of control accuracy, reaction time and information transfer rate. We credit this success to the applied user-centered design approach and to a highly flexible technical setup. State-of-the art machine learning methods allowed the exploitation and combination of multiple relevant features contained in the EEG, which rapidly enabled the patients to gain substantial BCI control. Thus, we could show the feasibility of a flexible and tailorable BCI application in severely disabled users. This can be considered a significant success for two reasons: Firstly, the results were obtained within a short period of time, matching the tight clinical requirements. Secondly, the participating patients showed, compared to most other studies, very severe communication deficits. They were dependent on everyday use of AT and two patients were in a locked-in state. For the most affected patient a reliable communication was rarely possible with existing AT.

Recently, spatio-temporal filtering to enhance decoding for Brain-Computer-Interfacing (BCI) has become increasingly popular. In this work, we discuss a novel, fully Bayesian-and thereby probabilistic-framework, called Bayesian Spatio-Spectral Filter Optimization (BSSFO) and apply it to a large data set of 80 non-invasive EEG-based BCI experiments. Across the full frequency range, the BSSFO framework allows to analyze which spatio-spectral parameters are common and which ones differ across the subject population. As expected, large variability of brain rhythms is observed between subjects. We have clustered subjects according to similarities in their corresponding spectral characteristics from the BSSFO model, which is found to reflect their BCI performances well. In BCI, a considerable percentage of subjects is unable to use a BCI for communication, due to their missing ability to modulate their brain rhythms-a phenomenon sometimes denoted as BCI-illiteracy or inability. Predicting individual subjects' performance preceding the actual, time-consuming BCI-experiment enhances the usage of BCIs, e.g., by detecting users with BCI inability. This work additionally contributes by using the novel BSSFO method to predict the BCI-performance using only 2 minutes and 3 channels of resting-state EEG data recorded before the actual BCI-experiment. Specifically, by grouping the individual frequency characteristics we have nicely classified them into the subject 'prototypes' (like μ - or β -rhythm type subjects) or users without ability to communicate with a BCI, and then by further building a linear regression model based on the grouping we could predict subjects' performance with the maximum correlation coefficient of 0.581 with the performance later seen in the actual BCI session.

Event-related potential (ERP)-based P300 spellers are commonly used in the field of brain-computer interfaces as an alternative channel of communication for people with severe neuro-muscular diseases. This study introduces a novel P300 based brain-computer interface (BCI) stimulus paradigm using a random set presentation pattern and exploiting the effects of face familiarity. The effect of face familiarity is widely studied in the cognitive neurosciences and has recently been addressed for the purpose of BCI. In this study we compare P300-based BCI performances of a conventional row-column (RC)-based paradigm with our approach that combines a random set presentation paradigm with (non-) self-face stimuli. Our experimental results indicate stronger deflections of the ERPs in response to face stimuli, which are further enhanced when using the self-face images, and thereby improving P300-based spelling performance. This lead to a significant reduction of stimulus sequences required for correct character classification. These findings demonstrate a promising new approach for improving the speed and thus fluency of BCI-enhanced communication with the widely used P300-based BCI setup.

Despite several approaches to realize subject-to-subject transfer of pre-trained classifiers, the full performance of a Brain-Computer Interface (BCI) for a novel user can only be reached by presenting the BCI system with data from the novel user. In typical state-of-the-art BCI systems with a supervised classifier, the labeled data is collected during a calibration recording, in which the user is asked to perform a specific task. Based on the known labels of this recording, the BCI's classifier can learn to decode the individual's brain signals. Unfortunately, this calibration recording consumes valuable time. Furthermore, it is unproductive with respect to the final BCI application, e.g. text entry. Therefore, the calibration period must be reduced to a minimum, which is especially important for patients with a limited concentration ability. The main contribution of this manuscript is an online study on unsupervised learning in an auditory event-related potential (ERP) paradigm. Our results demonstrate that the calibration recording can be bypassed by utilizing an unsupervised trained classifier, that is initialized randomly and updated during usage. Initially, the unsupervised classifier tends to make decoding mistakes, as the classifier might not have seen enough data to build a reliable model. Using a constant re-analysis of the previously spelled symbols, these initially misspelled symbols can be rectified posthoc when the classifier has learned to decode the signals. We compare the spelling performance of our unsupervised approach and of the unsupervised posthoc approach to the standard supervised calibration-based dogma for n = 10 healthy users. To assess the learning behavior of our approach, it is unsupervised trained from scratch three times per user. Even with the relatively low SNR of an auditory ERP paradigm, the results show that after a limited number of trials (30 trials), the unsupervised approach performs comparably to a classic supervised model.

Robust and reliable covariance estimates play a decisive role in financial and many other applications. An important class of estimators is based on factor models. Here, we show by extensive Monte Carlo simulations that covariance matrices derived from the statistical Factor Analysis model exhibit a systematic error, which is similar to the well-known systematic error of the spectrum of the sample covariance matrix. Moreover, we introduce the Directional Variance Adjustment (DVA) algorithm, which diminishes the systematic error. In a thorough empirical study for the US, European, and Hong Kong stock market we show that our proposed method leads to improved portfolio allocation.