Search

Characterizing the subtle changes of functional brain networks associated with the pathological cascade of Alzheimer's disease (AD) is important for early diagnosis and prediction of disease progression prior to clinical symptoms. We developed a new deep learning method, termed multiple graph Gaussian embedding model (MG2G), which can learn highly informative network features by mapping high-dimensional resting-state brain networks into a low-dimensional latent space. These latent distribution-based embeddings enable a quantitative characterization of subtle and heterogeneous brain connectivity patterns at different regions and can be used as input to traditional classifiers for various downstream graph analytic tasks, such as AD early stage prediction, and statistical evaluation of between-group significant alterations across brain regions. We used MG2G to detect the intrinsic latent dimensionality of MEG brain networks, predict the progression of patients with mild cognitive impairment (MCI) to AD, and identify brain regions with network alterations related to MCI.

Humans perform remarkably well in many cognitive tasks including pattern recognition. However, the neuronal mechanisms underlying this process are not well understood. Nevertheless, artificial neural networks, inspired in brain circuits, have been designed and used to tackle spatio-temporal pattern recognition tasks. In this paper we present a multineuronal spike pattern detection structure able to autonomously implement online learning and recognition of parallel spike sequences (i.e., sequences of pulses belonging to different neurons/neural ensembles). The operating principle of this structure is based on two spiking/synaptic neurocomputational characteristics: spike latency, that enables neurons to fire spikes with a certain delay and heterosynaptic plasticity, that allows the own regulation of synaptic weights. From the perspective of the information representation, the structure allows mapping a spatio-temporal stimulus into a multidimensional, temporal, feature space. In this space, the parameter coordinate and the time at which a neuron fires represent one specific feature. In this sense, each feature can be considered to span a single temporal axis. We applied our proposed scheme to experimental data obtained from a motor inhibitory cognitive task. The test exhibits good classification performance, indicating the adequateness of our approach. In addition to its effectiveness, its simplicity and low computational cost suggest a large scale implementation for real time recognition applications in several areas, such as brain computer interface, personal biometrics authentication or early detection of diseases., Comment: Submitted to Frontiers in Neuroscience

Limitations in processing capabilities and memory of today's computers make spiking neuron-based (human) whole-brain simulations inevitably characterized by a compromise between bio-plausibility and computational cost. It translates into brain models composed of a reduced number of neurons and a simplified neuron's mathematical model, leading to the search for new simulation strategies. Taking advantage of the sparse character of brain-like computation, the event-driven technique could represent a way to carry out efficient simulation of large-scale Spiking Neural Networks (SNN). The recent Leaky Integrate-and-Fire with Latency (LIFL) spiking neuron model is event-driven compatible and exhibits some realistic neuronal features, opening new avenues for brain modelling. In this paper we introduce FNS, the first LIFL-based spiking neural network framework, which combines spiking/synaptic neural modelling with the event-driven approach, allowing us to define heterogeneous neuron modules and multi-scale connectivity with delayed connections and plastic synapses. In order to allow multi-thread implementations a novel parallelization strategy is also introduced. This paper presents mathematical models, software implementation and simulation routines on which FNS is based. Finally, a brain subnetwork is modeled on the basis of real brain structural data, and the resulting simulated activity is compared with associated brain functional (source-space MEG) data, demonstrating a good matching between the activity of the model and that of the experimetal data. This work aims to lay the groundwork for future event-driven based personalised brain models., Comment: Changed title, added references, corrected typos

We investigated how the organization of functional brain networks was related to cognitive reserve (CR) during a memory task in healthy aging. We obtained the magnetoencephalographic functional networks of 20 elders with a high or low CR level to analyse the differences at network features. We reported a negative correlation between synchronization of the whole network and CR, and observed differences both at the node and at the network level in: the average shortest path and the network outreach. Individuals with high CR required functional networks with lower links to successfully carry out the memory task. These results may indicate that those individuals with low CR level exhibited a dual pattern of compensation and network impairment, since their functioning was more energetically costly to perform the task as the high CR group. Additionally, we evaluated how the dynamical properties of the different brain regions were correlated to the network parameters obtaining that entropy was positively correlated with the strength and clustering coefficient, while complexity behaved conversely. Consequently, highly connected nodes of the functional networks showed a more stochastic and less complex signal. We consider that network approach may be a relevant tool to better understand brain functioning in aging., Comment: Main manuscript: 23 pages including references, 20 pages text, 8 figures and supplementary information included

Despite the increase in calculation power in the last decades, the estimation of brain connectivity is still a tedious task. The high computational cost of the algorithms escalates with the square of the number of signals evaluated, usually in the range of thousands. In this work we propose a re-formulation of a widely used algorithm that allows the estimation of whole brain connectivity in much smaller times. We start from the original implementation of Phase Locking Value (PLV) and re-formulated it in a highly computational efficient way. Besides, this formulation stresses its strong similarity with coherence, which we used to introduce two new metrics insensitive to zero lag synchronization, the imaginary part of PLV (iPLV) and its corrected counterpart (ciPLV). The new implementation of PLV avoids some highly CPU-expensive operations, and achieved a 100-fold speedup over the original algorithm. The new derived metrics were highly robust in the presence of volume conduction. ciPLV, in particular, proved capable of ignoring zero-lag connectivity, while correctly estimating nonzero-lag connectivity. Our implementation of PLV makes it possible to calculate whole-brain connectivity in much shorter times. The results of the simulations using ciPLV suggest that this metric is ideal to measure synchronization in the presence of volume conduction or source leakage effects.

We investigate how hubs of functional brain networks are modified as a result of mild cognitive impairment (MCI), a condition causing a slight but noticeable decline in cognitive abilities, which sometimes precedes the onset of Alzheimer's disease. We used magnetoencephalography (MEG) to investigate the functional brain networks of a group of patients suffering from MCI and a control group of healthy subjects, during the execution of a short-term memory task. Couplings between brain sites were evaluated using synchronization likelihood, from which a network of functional interdependencies was constructed and the centrality, i.e. importance, of their nodes quantified. The results showed that, with respect to healthy controls, MCI patients were associated with decreases and increases in hub centrality respectively in occipital and central scalp regions, supporting the hypothesis that MCI modifies functional brain network topology, leading to more random structures., Comment: 12 pages, 5 figures, to appear in International Journal of Bifurcations and Chaos, 2013

The analysis of the interdependence between time series has become an important field of research in the last years, mainly as a result of advances in the characterization of dynamical systems from the signals they produce, the introduction of concepts such as generalized and phase synchronization and the application of information theory to time series analysis. In neurophysiology, different analytical tools stemming from these concepts have added to the 'traditional' set of linear methods, which includes the cross-correlation and the coherency function in the time and frequency domain, respectively, or more elaborated tools such as Granger Causality. This increase in the number of approaches to tackle the existence of functional (FC) or effective connectivity (EC) between two (or among many) neural networks, along with the mathematical complexity of the corresponding time series analysis tools, makes it desirable to arrange them into a unified-easy-to-use software package. The goal is to allow neuroscientists, neurophysiologists and researchers from related fields to easily access and make use of these analysis methods from a single integrated toolbox. Here we present HERMES (http://hermes.ctb.upm.es), a toolbox for the Matlab environment (The Mathworks, Inc), which is designed for the analysis functional and effective brain connectivity from neurophysiological data such as multivariate EEG and/or MEG records. It includes also visualization tools and statistical methods to address the problem of multiple comparisons. We believe that this toolbox will be very helpful to all the researchers working in the emerging field of brain connectivity analysis., Comment: 58 pages, 10 figures, 3 tables, Neuroinformatics 2013