Pierre Yger

An Optimized Blockwise Nonlocal Means Denoising Filter for 3-D Magnetic Resonance Images

A critical issue in image restoration is the problem of noise removal while keeping the integrity of relevant image information. Denoising is a crucial step to increase image quality and to improve the performance of all the tasks needed for quantitative imaging analysis. The method proposed in this paper is based on a 3-D optimized blockwise version of the nonlocal (NL)-means filter (Buades, , 2005). The NL-means filter uses the redundancy of information in the image under study to remove the noise. The performance of the NL-means filter has been already demonstrated for 2-D images, but reducing the computational burden is a critical aspect to extend the method to 3-D images. To overcome this problem, we propose improvements to reduce the computational complexity. These different improvements allow to drastically divide the computational time while preserving the performances of the NL-means filter. A fully automated and optimized version of the NL-means filter is then presented. Our contributions to the NL-means filter are: 1) an automatic tuning of the smoothing parameter; 2) a selection of the most relevant voxels; 3) a blockwise implementation; and 4) a parallelized computation. Quantitative validation was carried out on synthetic datasets generated with BrainWeb (Collins, , 1998). The results show that our optimized NL-means filter outperforms the classical implementation of the NL-means filter, as well as two other classical denoising methods [anisotropic diffusion (Perona and Malik, 1990)] and total variation minimization process (Rudin, , 1992) in terms of accuracy (measured by the peak signal-to-noise ratio) with low computation time. Finally, qualitative results on real data are presented.

PyNN: A Common Interface for Neuronal Network Simulators

Computational neuroscience has produced a diversity of software for simulations of networks of spiking neurons, with both negative and positive consequences. On the one hand, each simulator uses its own programming or configuration language, leading to considerable difficulty in porting models from one simulator to another. This impedes communication between investigators and makes it harder to reproduce and build on the work of others. On the other hand, simulation results can be cross-checked between different simulators, giving greater confidence in their correctness, and each simulator has different optimizations, so the most appropriate simulator can be chosen for a given modelling task. A common programming interface to multiple simulators would reduce or eliminate the problems of simulator diversity while retaining the benefits. PyNN is such an interface, making it possible to write a simulation script once, using the Python programming language, and run it without modification on any supported simulator (currently NEURON, NEST, PCSIM, Brian and the Heidelberg VLSI neuromorphic hardware). PyNN increases the productivity of neuronal network modelling by providing high-level abstraction, by promoting code sharing and reuse, and by providing a foundation for simulator-agnostic analysis, visualization and data-management tools. PyNN increases the reliability of modelling studies by making it much easier to check results on multiple simulators. PyNN is open-source software and is available from http://neuralensemble.org/PyNN.

Fast non local means denoising for 3d MR images

One critical issue in the context of image restoration is the problem of noise removal while keeping the integrity of relevant image information. Denoising is a crucial step to increase image conspicuity and to improve the performances of all the processings needed for quantitative imaging analysis. The method proposed in this paper is based on an optimized version of the Non Local (NL) Means algorithm. This approach uses the natural redundancy of information in image to remove the noise. Tests were carried out on synthetic datasets and on real 3T MR images. The results show that the NL-means approach outperforms other classical denoising methods, such as Anisotropic Diffusion Filter and Total Variation.

Network-state modulation of power-law frequency-scaling in visual cortical neurons.

Various types of neural-based signals, such as EEG, local field potentials and intracellular synaptic potentials, integrate multiple sources of activity distributed across large assemblies. They have in common a power-law frequency-scaling structure at high frequencies, but it is still unclear whether this scaling property is dominated by intrinsic neuronal properties or by network activity. The latter case is particularly interesting because if frequency-scaling reflects the network state it could be used to characterize the functional impact of the connectivity. In intracellularly recorded neurons of cat primary visual cortex in vivo, the power spectral density of Vm activity displays a power-law structure at high frequencies with a fractional scaling exponent. We show that this exponent is not constant, but depends on the visual statistics used to drive the network. To investigate the determinants of this frequency-scaling, we considered a generic recurrent model of cortex receiving a retinotopically organized external input. Similarly to the in vivo case, our in computo simulations show that the scaling exponent reflects the correlation level imposed in the input. This systematic dependence was also replicated at the single cell level, by controlling independently, in a parametric way, the strength and the temporal decay of the pairwise correlation between presynaptic inputs. This last model was implemented in vitro by imposing the correlation control in artificial presynaptic spike trains through dynamic-clamp techniques. These in vitro manipulations induced a modulation of the scaling exponent, similar to that observed in vivo and predicted in computo. We conclude that the frequency-scaling exponent of the Vm reflects stimulus-driven correlations in the cortical network activity. Therefore, we propose that the scaling exponent could be used to read-out the “effective” connectivity responsible for the dynamical signature of the population signals measured at different integration levels, from Vm to LFP, EEG and fMRI.

Neo: an object model for handling electrophysiology data in multiple formats

Neuroscientists use many different software tools to acquire, analyze and visualize electrophysiological signals. However, incompatible data models and file formats make it difficult to exchange data between these tools. This reduces scientific productivity, renders potentially useful analysis methods inaccessible and impedes collaboration between labs. A common representation of the core data would improve interoperability and facilitate data-sharing. To that end, we propose here a language-independent object model, named “Neo,” suitable for representing data acquired from electroencephalographic, intracellular, or extracellular recordings, or generated from simulations. As a concrete instantiation of this object model we have developed an open source implementation in the Python programming language. In addition to representing electrophysiology data in memory for the purposes of analysis and visualization, the Python implementation provides a set of input/output (IO) modules for reading/writing the data from/to a variety of commonly used file formats. Support is included for formats produced by most of the major manufacturers of electrophysiology recording equipment and also for more generic formats such as MATLAB. Data representation and data analysis are conceptually separate: it is easier to write robust analysis code if it is focused on analysis and relies on an underlying package to handle data representation. For that reason, and also to be as lightweight as possible, the Neo object model and the associated Python package are deliberately limited to representation of data, with no functions for data analysis or visualization. Software for neurophysiology data analysis and visualization built on top of Neo automatically gains the benefits of interoperability, easier data sharing and automatic format conversion; there is already a burgeoning ecosystem of such tools. We intend that Neo should become the standard basis for Python tools in neurophysiology.

Reliable Recall of Spontaneous Activity Patterns in Cortical Networks

Irregular ongoing activity in cortical networks is often modeled as arising from recurrent connectivity. Yet it remains unclear to what extent its presence corrupts sensory signal transmission and network computational capabilities. In a recurrent cortical-like network, we have determined the activity patterns that are better transmitted and self-sustained by the network. We show that reproducible spiking and subthreshold dynamics can be triggered if the statistics of the imposed external drive are consistent with patterns previously seen in the ongoing activity. A subset of neurons in the network, constrained to replay temporal pattern segments extracted from the recorded ongoing activity of the same network, reliably drives the remaining, free-running neurons to call the rest of the pattern. Comparison with surrogate Poisson patterns indicates that the efficiency of the recall and completion process depends on the similarity between the statistical properties of the input with previous ongoing activity The reliability of evoked dynamics in recurrent networks is thus dependent on the stimulus used, and we propose that the similarity between spontaneous and evoked activity in sensory cortical areas could be a signature of efficient transmission and propagation across cortical networks.

Topologically invariant macroscopic statistics in balanced networks of conductance-based integrate-and-fire neurons

The relationship between the dynamics of neural networks and their patterns of connectivity is far from clear, despite its importance for understanding functional properties. Here, we have studied sparsely-connected networks of conductance-based integrate-and-fire (IF) neurons with balanced excitatory and inhibitory connections and with finite axonal propagation speed. We focused on the genesis of states with highly irregular spiking activity and synchronous firing patterns at low rates, called slow Synchronous Irregular (SI) states. In such balanced networks, we examined the “macroscopic” properties of the spiking activity, such as ensemble correlations and mean firing rates, for different intracortical connectivity profiles ranging from randomly connected networks to networks with Gaussian-distributed local connectivity. We systematically computed the distance-dependent correlations at the extracellular (spiking) and intracellular (membrane potential) levels between randomly assigned pairs of neurons. The main finding is that such properties, when they are averaged at a macroscopic scale, are invariant with respect to the different connectivity patterns, provided the excitatory-inhibitory balance is the same. In particular, the same correlation structure holds for different connectivity profiles. In addition, we examined the response of such networks to external input, and found that the correlation landscape can be modulated by the mean level of synchrony imposed by the external drive. This modulation was found again to be independent of the external connectivity profile. We conclude that first and second-order “mean-field” statistics of such networks do not depend on the details of the connectivity at a microscopic scale. This study is an encouraging step toward a mean-field description of topological neuronal networks.

Stable Learning in Stochastic Network States

The mammalian cerebral cortex is characterized in vivo by irregular spontaneous activity, but how this ongoing dynamics affects signal processing and learning remains unknown. The associative plasticity rules demonstrated in vitro, mostly in silent networks, are based on the detection of correlations between presynaptic and postsynaptic activity and hence are sensitive to spontaneous activity and spurious correlations. Therefore, they cannot operate in realistic network states. Here, we present a new class of spike-timing-dependent plasticity learning rules with local floating plasticity thresholds, the slow dynamics of which account for metaplasticity. This novel algorithm is shown to both correctly predict homeostasis in synaptic weights and solve the problem of asymptotic stable learning in noisy states. It is shown to naturally encompass many other known types of learning rule, unifying them into a single coherent framework. The mixed presynaptic and postsynaptic dependency of the floating plasticity threshold is justified by a cascade of known molecular pathways, which leads to experimentally testable predictions.

Fast and accurate spike sorting in vitro and in vivo for up to thousands of electrodes

Understanding how assemblies of neurons encode information requires recording large populations of cells in the brain. In recent years, multi-electrode arrays and large silicon probes have been developed to record simultaneously from hundreds or thousands of electrodes packed with a high density. However, these new devices challenge the classical way to do spike sorting. Here we developed a new method to solve these issues, based on a highly automated algorithm to extract spikes from extracellular data, and show that this algorithm reached near optimal performance both in vitro and in vivo. The algorithm is composed of two main steps: 1) a “template-finding” phase to extract the cell templates, i.e. the pattern of activity evoked over many electrodes when one neuron fires an action potential; 2) a “template-matching” phase where the templates were matched to the raw data to find the location of the spikes. The manual intervention by the user was reduced to the minimal, and the time spent on manual curation did not scale with the number of electrodes. We tested our algorithm with large-scale data from in vitro and in vivo recordings, from 32 to 4225 electrodes. We performed simultaneous extracellular and patch recordings to obtain “ground truth” data, i.e. cases where the solution to the sorting problem is at least partially known. The performance of our algorithm was always close to the best expected performance. We thus provide a general solution to sort spikes from large-scale extracellular recordings.

Evaluation of a new optimisation algorithm for rigid registration of MRI data

We propose to use a recently introduced optimisation method in the context of rigid registration of medical images. This optimisation method, introduced by Powell and called NEWUOA, is compared with two other widely used algorithms: Powells direction set and Nelder-Meads downhill simplex method. This paper performs a comparative evaluation of the performances of these algorithms to optimise different image similarity measures for different mono- and multi-modal registrations. Images from the BrainWeb project are used as a gold standard for validation purposes. This paper exhibits that the proposed optimisation algorithm is more robust, more accurate and faster than the two other methods.