Exploring the influence of functional architecture on cortical thickness networks in early psychosis - A longitudinal study.

Cortical thickness reductions differ between individuals with psychotic disorders and comparison subjects even in early stages of illness. Whether these reductions covary as expected by functional network membership or simply by spatial proximity has not been fully elucidated. Through orthonormal projective non-negative matrix factorization, cortical thickness measurements in functionally-annotated regions from MRI scans of early-stage psychosis and matched healthy controls were reduced in dimensionality into features capturing positive covariance. Rather than matching the functional networks, the covarying regions in each feature displayed a more localized spatial organization. With Bayesian belief networks, the covarying regions per feature were arranged into a network topology to visualize the dependency structure and identify key driving regions. The features demonstrated diagnosis-specific differences in cortical thickness distributions per feature, identifying reduction-vulnerable spatial regions. Differences in key cortical thickness features between psychosis and control groups were delineated, as well as those between affective and non-affective psychosis. Clustering of the participants, stratified by diagnosis and clinical variables, characterized the clinical traits that define the cortical thickness patterns. Longitudinal follow-up revealed that in select clusters with low baseline cortical thickness, clinical traits improved over time. Our study represents a novel effort to characterize brain structure in relation to functional networks in healthy and clinical populations and to map patterns of cortical thickness alterations among ESP patients onto clinical variables for a better understanding of brain pathophysiology.

Mechanical Property Based Brain Age Prediction using Convolutional Neural Networks.

Brain age is a quantitative estimate to explain an individual's structural and functional brain measurements relative to the overall population and is particularly valuable in describing differences related to developmental or neurodegenerative pathology. Accurately inferring brain age from brain imaging data requires sophisticated models that capture the underlying age-related brain changes. Magnetic resonance elastography (MRE) is a phase contrast MRI technology that uses external palpations to measure brain mechanical properties. Mechanical property measures of viscoelastic shear stiffness and damping ratio have been found to change across the entire life span and to reflect brain health due to neurodegenerative diseases and even individual differences in cognitive function. Here we develop and train a multi-modal 3D convolutional neural network (CNN) to model the relationship between age and whole brain mechanical properties. After training, the network maps the measurements and other inputs to a brain age prediction. We found high performance using the 3D maps of various mechanical properties to predict brain age. Stiffness maps alone were able to predict ages of the test group subjects with a mean absolute error (MAE) of 3.76 years, which is comparable to single inputs of damping ratio (MAE: 3.82) and outperforms single input of volume (MAE: 4.60). Combining stiffness and volume in a multimodal approach performed the best, with an MAE of 3.60 years, whereas including damping ratio worsened model performance. Our results reflect previous MRE literature that had demonstrated that stiffness is more strongly related to chronological age than damping ratio. This machine learning model provides the first prediction of brain age from brain biomechanical data-an advancement towards sensitively describing brain integrity differences in individuals with neuropathology.

Shift-invariant waveform learning on epileptic ECoG.

Seizure detection algorithms must discriminate abnormal neuronal activity associated with a seizure from normal neural activity in a variety of conditions. Our approach is to seek spatiotemporal waveforms with distinct morphology in electrocorticographic (ECoG) recordings of epileptic patients that are indicative of a subsequent seizure (preictal) versus non-seizure segments (interictal). To find these waveforms we apply a shift-invariant k-means algorithm to segments of spatially filtered signals to learn codebooks of prototypical waveforms. The frequency of the cluster labels from the codebooks is then used to train a binary classifier that predicts the class (preictal or interictal) of a test ECoG segment. We use the Matthews correlation coefficient to evaluate the performance of the classifier and the quality of the codebooks. We found that our method finds recurrent non-sinusoidal waveforms that could be used to build interpretable features for seizure prediction and that are also physiologically meaningful.

Women's health in The BMJ: a data science history.

OBJECTIVE: To determine how the representation of women's health has changed in clinical studies over the course of 70 years. DESIGN: Observational study of 71 866 research articles published between 1948 and 2018 in The BMJ. MAIN OUTCOME MEASURES: The incidence of women-specific health topics over time. General linear, additive and segmented regression models were used to estimate trends. RESULTS: Over 70 years, the overall odds that a word in a BMJ research article was 'woman' or 'women' increased by an annual factor of 1.023, but this rate of increase varied by clinical specialty with some showing little or no change. The odds that an article was about some aspect of women-specific health increased much more slowly, by an annual factor of 1.004. The incidence of articles about particular areas of women-specific medicine such as pregnancy did not show a general increase, but rather fluctuated over time. The incidence of articles making any mention of women, gender or sex declined between 1948 and 2005, after which it rose steeply so that by 2018 few papers made no mention of them at all. CONCLUSIONS: Over time women have become ever more prominent in BMJ research articles. However, the importance of women-specific health topics has waxed and waned as researchers responded ephemerally to medical advances, public health programmes, and sociolegal changes. The appointment of a woman editor-inchief in 2005 may have had a dramatic effect on whether women were mentioned in research articles.

Improving reference prioritisation with PICO recognition.

BACKGROUND: Machine learning can assist with multiple tasks during systematic reviews to facilitate the rapid retrieval of relevant references during screening and to identify and extract information relevant to the study characteristics, which include the PICO elements of patient/population, intervention, comparator, and outcomes. The latter requires techniques for identifying and categorising fragments of text, known as named entity recognition. METHODS: A publicly available corpus of PICO annotations on biomedical abstracts is used to train a named entity recognition model, which is implemented as a recurrent neural network. This model is then applied to a separate collection of abstracts for references from systematic reviews within biomedical and health domains. The occurrences of words tagged in the context of specific PICO contexts are used as additional features for a relevancy classification model. Simulations of the machine learning-assisted screening are used to evaluate the work saved by the relevancy model with and without the PICO features. Chi-squared and statistical significance of positive predicted values are used to identify words that are more indicative of relevancy within PICO contexts. RESULTS: Inclusion of PICO features improves the performance metric on 15 of the 20 collections, with substantial gains on certain systematic reviews. Examples of words whose PICO context are more precise can explain this increase. CONCLUSIONS: Words within PICO tagged segments in abstracts are predictive features for determining inclusion. Combining PICO annotation model into the relevancy classification pipeline is a promising approach. The annotations may be useful on their own to aid users in pinpointing necessary information for data extraction, or to facilitate semantic search.

Quantifying risk factors in medical reports with a context-aware linear model.

OBJECTIVE: We seek to quantify the mortality risk associated with mentions of medical concepts in textual electronic health records (EHRs). Recognizing mentions of named entities of relevant types (eg, conditions, symptoms, laboratory tests or behaviors) in text is a well-researched task. However, determining the level of risk associated with them is partly dependent on the textual context in which they appear, which may describe severity, temporal aspects, quantity, etc. METHODS: To take into account that a given word appearing in the context of different risk factors (medical concepts) can make different contributions toward risk level, we propose a multitask approach, called context-aware linear modeling, which can be applied using appropriately regularized linear regression. To improve the performance for risk factors unseen in training data (eg, rare diseases), we take into account their distributional similarity to other concepts. RESULTS: The evaluation is based on a corpus of 531 reports from EHRs with 99 376 risk factors rated manually by experts. While context-aware linear modeling significantly outperforms single-task models, taking into account concept similarity further improves performance, reaching the level of human annotators' agreements. CONCLUSION: Our results show that automatic quantification of risk factors in EHRs can achieve performance comparable to human assessment, and taking into account the multitask structure of the problem and the ability to handle rare concepts is crucial for its accuracy.

Prioritising references for systematic reviews with RobotAnalyst: A user study.

Screening references is a time-consuming step necessary for systematic reviews and guideline development. Previous studies have shown that human effort can be reduced by using machine learning software to prioritise large reference collections such that most of the relevant references are identified before screening is completed. We describe and evaluate RobotAnalyst, a Web-based software system that combines text-mining and machine learning algorithms for organising references by their content and actively prioritising them based on a relevancy classification model trained and updated throughout the process. We report an evaluation over 22 reference collections (most are related to public health topics) screened using RobotAnalyst with a total of 43 610 abstract-level decisions. The number of references that needed to be screened to identify 95% of the abstract-level inclusions for the evidence review was reduced on 19 of the 22 collections. Significant gains over random sampling were achieved for all reviews conducted with active prioritisation, as compared with only two of five when prioritisation was not used. RobotAnalyst's descriptive clustering and topic modelling functionalities were also evaluated by public health analysts. Descriptive clustering provided more coherent organisation than topic modelling, and the content of the clusters was apparent to the users across a varying number of clusters. This is the first large-scale study using technology-assisted screening to perform new reviews, and the positive results provide empirical evidence that RobotAnalyst can accelerate the identification of relevant studies. The results also highlight the issue of user complacency and the need for a stopping criterion to realise the work savings.

A semi-supervised approach using label propagation to support citation screening.

Citation screening, an integral process within systematic reviews that identifies citations relevant to the underlying research question, is a time-consuming and resource-intensive task. During the screening task, analysts manually assign a label to each citation, to designate whether a citation is eligible for inclusion in the review. Recently, several studies have explored the use of active learning in text classification to reduce the human workload involved in the screening task. However, existing approaches require a significant amount of manually labelled citations for the text classification to achieve a robust performance. In this paper, we propose a semi-supervised method that identifies relevant citations as early as possible in the screening process by exploiting the pairwise similarities between labelled and unlabelled citations to improve the classification performance without additional manual labelling effort. Our approach is based on the hypothesis that similar citations share the same label (e.g., if one citation should be included, then other similar citations should be included also). To calculate the similarity between labelled and unlabelled citations we investigate two different feature spaces, namely a bag-of-words and a spectral embedding based on the bag-of-words. The semi-supervised method propagates the classification codes of manually labelled citations to neighbouring unlabelled citations in the feature space. The automatically labelled citations are combined with the manually labelled citations to form an augmented training set. For evaluation purposes, we apply our method to reviews from clinical and public health. The results show that our semi-supervised method with label propagation achieves statistically significant improvements over two state-of-the-art active learning approaches across both clinical and public health reviews.

Eliciting naturalistic cortical responses with a sensory prosthesis via optimized microstimulation.

OBJECTIVE: Lost sensations, such as touch, could one day be restored by electrical stimulation along the sensory neural pathways. Such stimulation, when informed by electronic sensors, could provide naturalistic cutaneous and proprioceptive feedback to the user. Perceptually, microstimulation of somatosensory brain regions produces localized, modality-specific sensations, and several spatiotemporal parameters have been studied for their discernibility. However, systematic methods for encoding a wide array of naturally occurring stimuli into biomimetic percepts via multi-channel microstimulation are lacking. More specifically, generating spatiotemporal patterns for explicitly evoking naturalistic neural activation has not yet been explored. APPROACH: We address this problem by first modeling the dynamical input-output relationship between multichannel microstimulation and downstream neural responses, and then optimizing the input pattern to reproduce naturally occurring touch responses as closely as possible. MAIN RESULTS: Here we show that such optimization produces responses in the S1 cortex of the anesthetized rat that are highly similar to natural, tactile-stimulus-evoked counterparts. Furthermore, information on both pressure and location of the touch stimulus was found to be highly preserved. SIGNIFICANCE: Our results suggest that the currently presented stimulus optimization approach holds great promise for restoring naturalistic levels of sensation.

Learning Recurrent Waveforms Within EEGs.

GOAL: We demonstrate an algorithm to automatically learn the time-limited waveforms associated with phasic events that repeatedly appear throughout an electroencephalogram. METHODS: To learn the phasic event waveforms we propose a multiscale modeling process that is based on existing shift-invariant dictionary learning algorithms. For each channel, waveforms at different temporal scales are learned based on the assumption that only a few waveforms occur in any window of the time-series, but the same waveforms reoccur throughout the signal. Once the waveforms are learned, the timing and amplitude of the phasic event occurrences are estimated using matching pursuit. To summarize the waveforms learned across multiple channels and subjects, we analyze their frequency content, their similarity to Gabor-Morlet wavelets, and perform shift-invariant k-means to cluster the waveforms. A prototype waveform from each cluster is then tested for differential spatial patterns between different motor imagery conditions. RESULTS: On multiple human EEG datasets, the learned waveforms capture key characteristics of signals they were trained to represent, with a consistency in waveform morphology and frequency content across multiple training sections and initializations. On multichannel datasets, the spatial amplitude patterns of the waveforms are also consistent and can be used to distinguish different modalities of motor imagery. CONCLUSION: We explored a methodology that can be used for modeling the recurrent waveforms in EEG traces. SIGNIFICANCE: The methodology automatically identifies the most frequent phasic event waveforms in EEG, which could then be used as features for automatic evaluation and comparison of EEG during sleep, pathology, or mentally engaging tasks.

Representing and decomposing neural potential signals.

This paper reviews methodologies for analyzing neural potentials via frequency, time-frequency, or wavelet representations, and adaptive models that estimate the signal's spatial or temporal structure. The fundamental assumptions of each method are discussed. In particular, the Fourier transform is contrasted with overcomplete representations, which are able to precisely delineate the timing and/or frequency of neural events. Finally, a novel approach that combines overcomplete representations with adaptive signal models is presented. This approach describes a continuous signal as a linear combination of reoccurring waveforms, referred to as phasic events, which are often associated with neural processing. The new methodology automatically learns the reoccurring waveforms and quantifies the neural potentials by the set of amplitudes and timings.

A tensor-product-kernel framework for multiscale neural activity decoding and control.

Brain machine interfaces (BMIs) have attracted intense attention as a promising technology for directly interfacing computers or prostheses with the brain's motor and sensory areas, thereby bypassing the body. The availability of multiscale neural recordings including spike trains and local field potentials (LFPs) brings potential opportunities to enhance computational modeling by enriching the characterization of the neural system state. However, heterogeneity on data type (spike timing versus continuous amplitude signals) and spatiotemporal scale complicates the model integration of multiscale neural activity. In this paper, we propose a tensor-product-kernel-based framework to integrate the multiscale activity and exploit the complementary information available in multiscale neural activity. This provides a common mathematical framework for incorporating signals from different domains. The approach is applied to the problem of neural decoding and control. For neural decoding, the framework is able to identify the nonlinear functional relationship between the multiscale neural responses and the stimuli using general purpose kernel adaptive filtering. In a sensory stimulation experiment, the tensor-product-kernel decoder outperforms decoders that use only a single neural data type. In addition, an adaptive inverse controller for delivering electrical microstimulation patterns that utilizes the tensor-product kernel achieves promising results in emulating the responses to natural stimulation.

Neural decoding with kernel-based metric learning.

In studies of the nervous system, the choice of metric for the neural responses is a pivotal assumption. For instance, a well-suited distance metric enables us to gauge the similarity of neural responses to various stimuli and assess the variability of responses to a repeated stimulus-exploratory steps in understanding how the stimuli are encoded neurally. Here we introduce an approach where the metric is tuned for a particular neural decoding task. Neural spike train metrics have been used to quantify the information content carried by the timing of action potentials. While a number of metrics for individual neurons exist, a method to optimally combine single-neuron metrics into multineuron, or population-based, metrics is lacking. We pose the problem of optimizing multineuron metrics and other metrics using centered alignment, a kernel-based dependence measure. The approach is demonstrated on invasively recorded neural data consisting of both spike trains and local field potentials. The experimental paradigm consists of decoding the location of tactile stimulation on the forepaws of anesthetized rats. We show that the optimized metrics highlight the distinguishing dimensions of the neural response, significantly increase the decoding accuracy, and improve nonlinear dimensionality reduction methods for exploratory neural analysis.

Joint optimization of algorithmic suites for EEG analysis.

Electroencephalogram (EEG) data analysis algorithms consist of multiple processing steps each with a number of free parameters. A joint optimization methodology can be used as a wrapper to fine-tune these parameters for the patient or application. This approach is inspired by deep learning neural network models, but differs because the processing layers for EEG are heterogeneous with different approaches used for processing space and time. Nonetheless, we treat the processing stages as a neural network and apply backpropagation to jointly optimize the parameters. This approach outperforms previous results on the BCI Competition II - dataset IV; additionally, it outperforms the common spatial patterns (CSP) algorithm on the BCI Competition III dataset IV. In addition, the optimized parameters in the architecture are still interpretable.

Information-theoretic metric learning: 2-D linear projections of neural data for visualization.

Intracortical neural recordings are typically high-dimensional due to many electrodes, channels, or units and high sampling rates, making it very difficult to visually inspect differences among responses to various conditions. By representing the neural response in a low-dimensional space, a researcher can visually evaluate the amount of information the response carries about the conditions. We consider a linear projection to 2-D space that also parametrizes a metric between neural responses. The projection, and corresponding metric, should preserve class-relevant information pertaining to different behavior or stimuli. We find the projection as a solution to the information-theoretic optimization problem of maximizing the information between the projected data and the class labels. The method is applied to two datasets using different types of neural responses: motor cortex neuronal firing rates of a macaque during a center-out reaching task, and local field potentials in the somatosensory cortex of a rat during tactile stimulation of the forepaw. In both cases, projected data points preserve the natural topology of targets or peripheral touch sites. Using the learned metric on the neural responses increases the nearest-neighbor classification rate versus the original data; thus, the metric is tuned to distinguish among the conditions.

Locating spatial patterns of waveforms during sensory perception in scalp EEG.

The spatio-temporal oscillations in EEG waves are indicative of sensory and cognitive processing. We propose a method to find the spatial amplitude patterns of a time-limited waveform across multiple EEG channels. It consists of a single iteration of multichannel matching pursuit where the base waveform is obtained via the Hilbert transform of a time-limited tone. The vector of extracted amplitudes across channels is used for classification, and we analyze the effect of deviation in temporal alignment of the waveform on classification performance. Results for a previously published dataset of 6 subjects show comparable results versus a more complicated criteria-based method.

Subspace matching thalamic microstimulation to tactile evoked potentials in rat somatosensory cortex.

We show experimental results that the evoked local field potentials of the rat somatosensory cortex from natural tactile touch of forepaw digits and matched thalamic microstimulation can be qualitatively and quantitively similar. In ongoing efforts to optimize the microstimulation settings (e.g., location, amplitude, etc.) to match the natural response, we investigate whether subspace projection methods, specifically the eigenface approach proposed in the computer vision community (Turk and Pentland 1991 [1]), can be used to choose the parameters of microstimulation such that the response matches a single tactile touch realization. Since the evoked potentials from multiple electrodes are high dimensional spatio-temporal data, the subspace projections improve computational efficiency and can reduce the effect of noisy realizations. In computing the PCA projections we use the peristimulus averages instead of the realizations. The dataset is pruned of unreliable stimulation types. A new subspace is computed for the pruned stimulation type, and is used to estimate a sequence of microstimulations to best match the natural responses. This microstimulation sequence is applied in vivo and quantitative analysis shows that per realization matching does statistically better than choosing randomly from the pruned subset.

An association framework to analyze dependence structure in time series.

The purpose of this paper is two-fold: first, to propose a modification to the generalized measure of association (GMA) framework that reduces the effect of temporal structure in time series; second, to assess the reliability of using association methods to capture dependence between pairs of EEG channels using their time series or envelopes. To achieve the first goal, the GMA algorithm was updated so as to minimize the effect of the correlation inherent in the time structure. The reliability of the modified scheme was then assessed on both synthetic and real data. Synthetic data was generated from a Clayton copula, for which null hypotheses of uncorrelatedness were constructed for the signal. The signal was processed such that the envelope emulated important characteristics of experimental EEG data. Results show that the modified GMA procedure can capture pairwise dependence between generated signals as well as their envelopes with good statistical power. Furthermore, applying GMA and Kendall's tau to quantify dependence using the extracted envelopes of processed EEG data concords with previous findings using the signal itself.

An electric field model for prediction of somatosensory (S1) cortical field potentials induced by ventral posterior lateral (VPL) thalamic microstimulation.

Microstimulation (MiSt) is used experimentally and clinically to activate localized populations of neural elements. However, it is difficult to predict-and subsequently control-neural responses to simultaneous current injection through multiple electrodes in an array. This is due to the unknown locations of neuronal elements in the extracellular medium that are excited by the superposition of multiple parallel current sources. We, therefore, propose a model that maps the computed electric field in the 3-D space surrounding the stimulating electrodes in one brain region to the local field potential (LFP) fluctuations evoked in a downstream region. Our model is trained with the recorded LFP waveforms in the primary somatosensory cortex (S1) resulting from MiSt applied in multiple electrode configurations in the ventral posterolateral nucleus (VPL) of the quiet awake rat. We then predict the cortical responses to MiSt in "novel" electrode configurations, a result that suggests that this technique could aid in the design of spatially optimized MiSt patterns through a multielectrode array.

Optimizing microstimulation using a reinforcement learning framework.

The ability to provide sensory feedback is desired to enhance the functionality of neuroprosthetics. Somatosensory feedback provides closed-loop control to the motor system, which is lacking in feedforward neuroprosthetics. In the case of existing somatosensory function, a template of the natural response can be used as a template of desired response elicited by electrical microstimulation. In the case of no initial training data, microstimulation parameters that produce responses close to the template must be selected in an online manner. We propose using reinforcement learning as a framework to balance the exploration of the parameter space and the continued selection of promising parameters for further stimulation. This approach avoids an explicit model of the neural response from stimulation. We explore a preliminary architecture--treating the task as a k-armed bandit--using offline data recorded for natural touch and thalamic microstimulation, and we examine the methods efficiency in exploring the parameter space while concentrating on promising parameter forms. The best matching stimulation parameters, from k = 68 different forms, are selected by the reinforcement learning algorithm consistently after 334 realizations.