A neural network enhanced generalized minimum variance self-tuning proportional, integral and derivative control algorithm for complex dynamic systems

A neural network enhanced proportional, integral and derivative (PID) controller is presented that combines the attributes of neural network learning with a generalized minimum-variance self-tuning control (STC) strategy. The neuro PID controller is structured with plant model identification and PID parameter tuning. The plants to be controlled are approximated by an equivalent model composed of a simple linear submodel to approximate plant dynamics around operating points, plus an error agent to accommodate the errors induced by linear submodel inaccuracy due to non-linearities and other complexities. A generalized recursive least-squares algorithm is used to identify the linear submodel, and a layered neural network is used to detect the error agent in which the weights are updated on the basis of the error between the plant output and the output from the linear submodel. The procedure for controller design is based on the equivalent model, and therefore the error agent is naturally functioned within the control law. In this way the controller can deal not only with a wide range of linear dynamic plants but also with those complex plants characterized by severe non-linearity, uncertainties and non-minimum phase behaviours. Two simulation studies are provided to demonstrate the effectiveness of the controller design procedure.

Identification of nonlinear systems with a dynamic recurrent neural network

Two approaches are presented to calculate the weights for a Dynamic Recurrent Neural Network (DRNN) in order to identify the input-output dynamics of a class of nonlinear systems. The number of states of the identified network is constrained to be the same as the number of states of the plant.

Inverse problems in neural field theory

We study inverse problems in neural field theory, i.e., the construction of synaptic weight kernels yielding a prescribed neural field dynamics. We address the issues of existence, uniqueness, and stability of solutions to the inverse problem for the Amari neural field equation as a special case, and prove that these problems are generally ill-posed. In order to construct solutions to the inverse problem, we first recast the Amari equation into a linear perceptron equation in an infinite-dimensional Banach or Hilbert space. In a second step, we construct sets of biorthogonal function systems allowing the approximation of synaptic weight kernels by a generalized Hebbian learning rule. Numerically, this construction is implemented by the Moore–Penrose pseudoinverse method. We demonstrate the instability of these solutions and use the Tikhonov regularization method for stabilization and to prevent numerical overfitting. We illustrate the stable construction of kernels by means of three instructive examples.

NeMo: a platform for neural modelling of spiking neurons using GPUs

Simulating spiking neural networks is of great interest to scientists wanting to model the functioning of the brain. However, large-scale models are expensive to simulate due to the number and interconnectedness of neurons in the brain. Furthermore, where such simulations are used in an embodied setting, the simulation must be real-time in order to be useful. In this paper we present NeMo, a platform for such simulations which achieves high performance through the use of highly parallel commodity hardware in the form of graphics processing units (GPUs). NeMo makes use of the Izhikevich neuron model which provides a range of realistic spiking dynamics while being computationally efficient. Our GPU kernel can deliver up to 400 million spikes per second. This corresponds to a real-time simulation of around 40 000 neurons under biologically plausible conditions with 1000 synapses per neuron and a mean firing rate of 10 Hz.

Connecting mean field models of neural activity to EEG and fMRI data

Progress in functional neuroimaging of the brain increasingly relies on the integration of data from complementary imaging modalities in order to improve spatiotemporal resolution and interpretability. However, the usefulness of merely statistical combinations is limited, since neural signal sources differ between modalities and are related non-trivially. We demonstrate here that a mean field model of brain activity can simultaneously predict EEG and fMRI BOLD with proper signal generation and expression. Simulations are shown using a realistic head model based on structural MRI, which includes both dense short-range background connectivity and long-range specific connectivity between brain regions. The distribution of modeled neural masses is comparable to the spatial resolution of fMRI BOLD, and the temporal resolution of the modeled dynamics, importantly including activity conduction, matches the fastest known EEG phenomena. The creation of a cortical mean field model with anatomically sound geometry, extensive connectivity, and proper signal expression is an important first step towards the model-based integration of multimodal neuroimages.

Understanding the effects of anesthetic agents on the EEG through neural field theory

Anesthetic and analgesic agents act through a diverse range of pharmacological mechanisms. Existing empirical data clearly shows that such "microscopic" pharmacological diversity is reflected in their "macroscopic" effects on the human electroencephalogram (EEG). Based on a detailed mesoscopic neural field model we theoretically posit that anesthetic induced EEG activity is due to selective parametric changes in synaptic efficacy and dynamics. Specifically, on the basis of physiologically constrained modeling, it is speculated that the selective modification of inhibitory or excitatory synaptic activity may differentially effect the EEG spectrum. Such results emphasize the importance of neural field theories of brain electrical activity for elucidating the principles whereby pharmacological agents effect the EEG. Such insights will contribute to improved methods for monitoring depth of anesthesia using the EEG.

A dynamic causal model of the coupling between pulse stimulation and neural activity

We present a dynamic causal model that can explain context-dependent changes in neural responses, in the rat barrel cortex, to an electrical whisker stimulation at different frequencies. Neural responses were measured in terms of local field potentials. These were converted into current source density (CSD) data, and the time series of the CSD sink was extracted to provide a time series response train. The model structure consists of three layers (approximating the responses from the brain stem to the thalamus and then the barrel cortex), and the latter two layers contain nonlinearly coupled modules of linear second-order dynamic systems. The interaction of these modules forms a nonlinear regulatory system that determines the temporal structure of the neural response amplitude for the thalamic and cortical layers. The model is based on the measured population dynamics of neurons rather than the dynamics of a single neuron and was evaluated against CSD data from experiments with varying stimulation frequency (1–40 Hz), random pulse trains, and awake and anesthetized animals. The model parameters obtained by optimization for different physiological conditions (anesthetized or awake) were significantly different. Following Friston, Mechelli, Turner, and Price (2000), this work is part of a formal mathematical system currently being developed (Zheng et al., 2005) that links stimulation to the blood oxygen level dependent (BOLD) functional magnetic resonance imaging (fMRI) signal through neural activity and hemodynamic variables. The importance of the model described here is that it can be used to invert the hemodynamic measurements of changes in blood flow to estimate the underlying neural activity.

Mathematical model for NF-kappaB-driven proliferation of adult neural stem cells

Neural stem cells (NSCs) are early precursors of neuronal and glial cells. NSCs are capable of generating identical progeny through virtually unlimited numbers of cell divisions (cell proliferation), producing daughter cells committed to differentiation. Nuclear factor kappa B (NF-kappaB) is an inducible, ubiquitous transcription factor also expressed in neurones, glia and neural stem cells. Recently, several pieces of evidence have been provided for a central role of NF-kappaB in NSC proliferation control. Here, we propose a novel mathematical model for NF-kappaB-driven proliferation of NSCs. We have been able to reconstruct the molecular pathway of activation and inactivation of NF-kappaB and its influence on cell proliferation by a system of nonlinear ordinary differential equations. Then we use a combination of analytical and numerical techniques to study the model dynamics. The results obtained are illustrated by computer simulations and are, in general, in accordance with biological findings reported by several independent laboratories. The model is able to both explain and predict experimental data. Understanding of proliferation mechanisms in NSCs may provide a novel outlook in both potential use in therapeutic approaches, and basic research as well.

Emergence of spatially heterogeneous burst suppression in a neural field model of electrocortical activity

Burst suppression in the electroencephalogram (EEG) is a well-described phenomenon that occurs during deep anesthesia, as well as in a variety of congenital and acquired brain insults. Classically it is thought of as spatially synchronous, quasi-periodic bursts of high amplitude EEG separated by low amplitude activity. However, its characterization as a “global brain state” has been challenged by recent results obtained with intracranial electrocortigraphy. Not only does it appear that burst suppression activity is highly asynchronous across cortex, but also that it may occur in isolated regions of circumscribed spatial extent. Here we outline a realistic neural field model for burst suppression by adding a slow process of synaptic resource depletion and recovery, which is able to reproduce qualitatively the empirically observed features during general anesthesia at the whole cortex level. Simulations reveal heterogeneous bursting over the model cortex and complex spatiotemporal dynamics during simulated anesthetic action, and provide forward predictions of neuroimaging signals for subsequent empirical comparisons and more detailed characterization. Because burst suppression corresponds to a dynamical end-point of brain activity, theoretically accounting for its spatiotemporal emergence will vitally contribute to efforts aimed at clarifying whether a common physiological trajectory is induced by the actions of general anesthetic agents. We have taken a first step in this direction by showing that a neural field model can qualitatively match recent experimental data that indicate spatial differentiation of burst suppression activity across cortex.

Exploration of the neural correlates of cerebral palsy for sensorimotor BCI control

Cerebral palsy (CP) includes a broad range of disorders, which can result in impairment of posture and movement control. Brain-computer interfaces (BCIs) have been proposed as assistive devices for individuals with CP. Better understanding of the neural processing underlying motor control in affected individuals could lead to more targeted BCI rehabilitation and treatment options. We have explored well-known neural correlates of movement, including event-related desynchronization (ERD), phase synchrony, and a recently-introduced measure of phase dynamics, in participants with CP and healthy control participants. Although present, significantly less ERD and phase locking were found in the group with CP. Additionally, inter-group differences in phase dynamics were also significant. Taken together these findings suggest that users with CP exhibit lower levels of motor cortex activation during motor imagery, as reflected in lower levels of ongoing mu suppression and less functional connectivity. These differences indicate that development of BCIs for individuals with CP may pose additional challenges beyond those faced in providing BCIs to healthy individuals.

Samba, a Xenopus hnRNP Expressed in Neural and Neural Crest Tissues

RNA binding proteins regulate gene expression at the posttranscriptional level and play important roles in embryonic development. Here, we report the cloning and expression of Samba, a Xenopus hnRNP that is maternally expressed and persists at least until tail bud stages. During gastrula stages, Samba is enriched in the dorsal regions. Subsequently, its expression is elevated only in neural and neural crest tissues. In the latter, Samba expression overlaps with that of Slug in migratory neural crest cells. Thereafter, Samba is maintained in the neural crest derivatives, as well as other neural tissues, including the anterior and posterior neural tube and the eyes. Overexpression of Samba in the animal pole leads to defects in neural crest migration and cranial cartilage development. Thus, Samba encodes a Xenopus hnRNP that is expressed early in neural and neural crest derivatives and may regulate crest cells migratory behavior. Developmental Dynamics 238:204-209, 2009. (C) 2008 Wiley-Liss, Inc.

Structural integrity identification based on smart materials and neural networks

This paper presents a non-model based technique to detect, locate, and characterize structural damage by combining the impedance-based structural health monitoring technique with an artificial neural network. The impedance-based structural health monitoring technique, which utilizes the electromechanical coupling property of piezoelectric materials, has shown engineering feasibility in a variety of practical field applications. Relying on high frequency structural excitations (typically>30 kHz), this technique is very sensitive to minor structural changes in the near field of the piezoelectric sensors. In order to quantitatively assess the state of structures, two sets of artificial neural networks, which utilize measured electrical impedance signals for input patterns, were developed. By employing high frequency ranges and by incorporating neural network features, this technique is able to detect the damage in its early stage and to estimate the nature of damage without prior knowledge of the model of structures. The paper concludes with an experimental example, an investigation on a massive quarter scale model of a steel bridge section, in order to verify the performance of this proposed methodology.

Non-destructive evaluation tool for monitoring and detection of structural damage by using neural network.

This work studies the capability of generalization of Neural Network using vibration based measurement data aiming at operating condition and health monitoring of mechanical systems. The procedure uses the backpropagation algorithm to classify the input patters of a system with different stiffness ratios. It has been investigated a large set of input data, containing various stiffness ratios as well as a reduced set containing only the extreme ones in order to study generalizing capability of the network. This allows to definition of Neural Networks capable to use a reduced set of data during the training phase. Once it is successfully trained, it could identify intermediate failure condition. Several conditions and intensities of damages have been studied by using numerical data. The Neural Network demonstrated a good capacity of generalization for all case. Finally, the proposal was tested with experimental data.

Artificial neural networks: A novel approach to analysing the nutritional ecology of a blowfly species, Chrysomya megacephala

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Structural health evaluation by optimization techinique and artificial neural network

This paper presents two different approaches to detect, locate, and characterize structural damage. Both techniques utilize electrical impedance in a first stage to locate the damaged area. In the second stage, to quantify the damage severity, one can use neural network, or optimization technique. The electrical impedance-based, which utilizes the electromechanical coupling property of piezoelectric materials, has shown engineering feasibility in a variety of practical field applications. Relying on high frequency structural excitations, this technique is very sensitive to minor structural changes in the near field of the piezoelectric sensors, and therefore, it is able to detect the damage in its early stage. Optimization approaches must be used for the case where a good condensed model is known, while neural network can be also used to estimate the nature of damage without prior knowledge of the model of the structure. The paper concludes with an experimental example in a welded cubic aluminum structure, in order to verify the performance of these two proposed methodologies.