Philip H. Goodman

Simulation of networks of spiking neurons: A review of tools and strategies

We review different aspects of the simulation of spiking neural networks. We start by reviewing the different types of simulation strategies and algorithms that are currently implemented. We next review the precision of those simulation strategies, in particular in cases where plasticity depends on the exact timing of the spikes. We overview different simulators and simulation environments presently available (restricted to those freely available, open source and documented). For each simulation tool, its advantages and pitfalls are reviewed, with an aim to allow the reader to identify which simulator is appropriate for a given task. Finally, we provide a series of benchmark simulations of different types of networks of spiking neurons, including Hodgkin–Huxley type, integrate-and-fire models, interacting with current-based or conductance-based synapses, using clock-driven or event-driven integration strategies. The same set of models are implemented on the different simulators, and the codes are made available. The ultimate goal of this review is to provide a resource to facilitate identifying the appropriate integration strategy and simulation tool to use for a given modeling problem related to spiking neural networks.

Heterogeneity in the pyramidal network of the medial prefrontal cortex

The prefrontal cortex is specially adapted to generate persistent activity that outlasts stimuli and is resistant to distractors, presumed to be the basis of working memory. The pyramidal network that supports this activity is unknown. Multineuron patch-clamp recordings in the ferret medial prefrontal cortex showed a heterogeneity of synapses interconnecting distinct subnetworks of different pyramidal cells. One subnetwork was similar to the pyramidal network commonly found in primary sensory areas, consisting of accommodating pyramidal cells interconnected with depressing synapses. The other subnetwork contained complex pyramidal cells with dual apical dendrites displaying nonaccommodating discharge patterns; these cells were hyper-reciprocally connected with facilitating synapses displaying pronounced synaptic augmentation and post-tetanic potentiation. These cellular, synaptic and network properties could amplify recurrent interactions between pyramidal neurons and support persistent activity in the prefrontal cortex.

Artificial neural networks improve the accuracy of cancer survival prediction

The TNM staging system originated as a response to the need for an accurate, consistent, universal cancer outcome prediction system. Since the TNM staging system was introduced in the 1950s, new prognostic factors have been identified and new methods for integrating prognostic factors have been developed. This study compares the prediction accuracy of the TNM staging system with that of artificial neural network statistical models.

Neuropeptide and calcium-binding protein gene expression profiles predict neuronal anatomical type in the juvenile rat

Neocortical neurones can be classified according to several independent criteria: morphological, physiological, and molecular expression (neuropeptides (NPs) and/or calcium‐binding proteins (CaBPs)). While it has been suggested that particular NPs and CaBPs characterize certain anatomical subtypes of neurones, there is also considerable overlap in their expression, and little is known about simultaneous expression of multiple NPs and CaBPs in morphologically characterized neocortical neurones. Here we determined the gene expression profiles of calbindin (CB), parvalbumin (PV), calretinin (CR), neuropeptide Y (NPY), vasoactive intestinal peptide (VIP), somatostatin (SOM) and cholecystokinin (CCK) in 268 morphologically identified neurones located in layers 2–6 in the juvenile rat somatosensory neocortex. We used patch‐clamp electrodes to label neurones with biocytin and harvest the cytoplasm to perform single‐cell RT‐multiplex PCR. Quality threshold clustering, an unsupervised algorithm that clustered neurones according to their entire profile of expressed genes, revealed seven distinct clusters. Surprisingly, each cluster preferentially contained one anatomical class. Artificial neural networks using softmax regression predicted anatomical types at nearly optimal statistical levels. Classification tree‐splitting (CART), a simple binary neuropeptide decision tree algorithm, revealed the manner in which expression of the multiple mRNAs relates to different anatomical classes. Pruning the CART tree revealed the key predictors of anatomical class (in order of importance: SOM, PV, VIP, and NPY). We reveal here, for the first time, a strong relationship between specific combinations of NP and CaBP gene expressions and the anatomical class of neocortical neurones.

Comparing artificial neural networks to other statistical methods for medical outcome prediction

Survival prediction is important in cancer because it determines therapy, matches patients for clinical trials, and provides patient information. Is a backpropagation neural network more accurate at predicting survival in breast cancer than the current staging system? For over thirty years cancer outcome prediction has been based on the pTNM staging system. There are two problems with this system: (1) it is not very accurate, and (2) its accuracy can not be improved because predictive variables can not be added to the model without increasing the models complexity to the point where it is no longer useful to the clinician. Using the area under the curve (AUC) of the receiver operating characteristic, the authors compare the accuracy of the following predictive models: pTNM stage, principal components analysis, classification and regression trees, logistic regression, cascade correlation neural network, conjugate gradient descent neural network, backpropagation neural network, and probabilistic neural network. Using just the TNM variables both the backpropagation neural network, AUC.768, and the probabilistic neural network, AUC.759, are significantly more accurate than the pTNM stage system, AUC.720 (all SEs<.01, p<.01 for both models compared to the pTNM model). Adding variables further increases the prediction accuracy of the backpropagation neural network, AUC.779, and the probabilistic neural network, AUC.777. Adding the new prognostic factors p53 and HER-2/neu increases the backpropagation neural networks accuracy to an AUC of .850. The neural networks perform equally well when applied to another breast cancer data set and to a colorectal cancer data set. Neural networks are able to significantly improve breast cancer outcome prediction accuracy when compared to the TNM stage system. They can combine prognostic factors to further improve accuracy. Neural networks are robust across data bases and cancer sites. Neural networks can perform as well as the best traditional prediction methods, and they can capture the power of nonmonotonic predictors and discover complex genetic interactions.<<ETX>>

Virtual neurorobotics (VNR) to accelerate development of plausible neuromorphic brain architectures

Traditional research in artificial intelligence and machine learning has viewed the brain as a specially adapted information-processing system. More recently the field of social robotics has been advanced to capture the important dynamics of human cognition and interaction. An overarching societal goal of this research is to incorporate the resultant knowledge about intelligence into technology for prosthetic, assistive, security, and decision support applications. However, despite many decades of investment in learning and classification systems, this paradigm has yet to yield truly “intelligent” systems. For this reason, many investigators are now attempting to incorporate more realistic neuromorphic properties into machine learning systems, encouraged by over two decades of neuroscience research that has provided parameters that characterize the brains interdependent genomic, proteomic, metabolomic, anatomic, and electrophysiological networks. Given the complexity of neural systems, developing tenable models to capture the essence of natural intelligence for real-time application requires that we discriminate features underlying information processing and intrinsic motivation from those reflecting biological constraints (such as maintaining structural integrity and transporting metabolic products). We propose herein a conceptual framework and an iterative method of virtual neurorobotics (VNR) intended to rapidly forward-engineer and test progressively more complex putative neuromorphic brain prototypes for their ability to support intrinsically intelligent, intentional interaction with humans. The VNR system is based on the viewpoint that a truly intelligent system must be driven by emotion rather than programmed tasking, incorporating intrinsic motivation and intentionality. We report pilot results of a closed-loop, real-time interactive VNR system with a spiking neural brain, and provide a video demonstration as online supplemental material.

A Component-Based Extension Framework for Large-Scale Parallel Simulations in NEURON.

As neuronal simulations approach larger scales with increasing levels of detail, the neurosimulator software represents only a part of a chain of tools ranging from setup, simulation, interaction with virtual environments to analysis and visualizations. Previously published approaches to abstracting simulator engines have not received wide-spread acceptance, which in part may be to the fact that they tried to address the challenge of solving the model specification problem. Here, we present an approach that uses a neurosimulator, in this case NEURON, to describe and instantiate the network model in the simulators native model language but then replaces the main integration loop with its own. Existing parallel network models are easily adopted to run in the presented framework. The presented approach is thus an extension to NEURON but uses a component-based architecture to allow for replaceable spike exchange components and pluggable components for monitoring, analysis, or control that can run in this framework alongside with the simulation.

The Accuracy of Artificial Neural Networks in Predicting Long-term Outcome After Traumatic Brain Injury

ObjectiveThis study compared the accuracy of artificial neural networks to multiple regression and classification and regression trees in predicting outcomes of 1644 patients in the Traumatic Brain Injury Model Systems database 1 year after injury. MethodsData from rehabilitation admission were used to predict discharge scores on the Functional Independence Measure, the Disability Rating Scale, and the Community Integration Questionnaire ResultsArtificial neural networks did not demonstrate greater accuracy in predicting outcomes than did the more widely used method of multiple regression. Both of these methods outperformed classification and regression trees ConclusionBecause of the sophisticated form of multiple regression with splines that was used, firm conclusions are limited about the relative accuracy of artificial neural networks compared to more widely used forms of multiple regression.

Brainlab: a Python toolkit to aid in the design, simulation, and analysis of spiking neural networks with the NeoCortical Simulator

Neuroscience modeling experiments often involve multiple complex neural network and cell model variants, complex input stimuli and input protocols, followed by complex data analysis. Coordinating all this complexity becomes a central difficulty for the experimenter. The Python programming language, along with its extensive library packages, has emerged as a leading “glue” tool for managing all sorts of complex programmatic tasks. This paper describes a toolkit called Brainlab, written in Python, that leverages Pythons strengths for the task of managing the general complexity of neuroscience modeling experiments. Brainlab was also designed to overcome the major difficulties of working with the NCS (NeoCortical Simulator) environment in particular. Brainlab is an integrated model-building, experimentation, and data analysis environment for the powerful parallel spiking neural network simulator system NCS.

Framework and implications of virtual neurorobotics

Despite decades of societal investment in artificial learning systems, truly “intelligent” systems have yet to be realized. These traditional models are based on input-output pattern optimization and/or cognitive production rule modeling. One response has been social robotics, using the interaction of human and robot to capture important cognitive dynamics such as cooperation and emotion; to date, these systems still incorporate traditional learning algorithms. More recently, investigators are focusing on the core assumptions of the brain “algorithm” itself—trying to replicate uniquely “neuromorphic” dynamics such as action potential spiking and synaptic learning. Only now are large-scale neuromorphic models becoming feasible, due to the availability of powerful supercomputers and an expanding supply of parameters derived from research into the brains interdependent electrophysiological, metabolomic and genomic networks. Personal computer technology has also led to the acceptance of computer-generated humanoid images, or “avatars”, to represent intelligent actors in virtual realities. In a recent paper, we proposed a method of virtual neurorobotics (VNR) in which the approaches above (social-emotional robotics, neuromorphic brain architectures, and virtual reality projection) are hybridized to rapidly forward-engineer and develop increasingly complex, intrinsically intelligent systems. In this paper, we synthesize our research and related work in the field and provide a framework for VNR, with wider implications for research and practical applications.