Evolutionary design of dynamic neural networks for evaporator control

This paper presents a novel neural network (NN) to control an ammonia refrigerant evaporator. Inspired by the latest findings on the biological neuron, a dynamic synaptic unit (DSU) is proposed to enhance the information processing capacity of artificial neurons. Treating the dynamic synaptic activity after the nonlinear somatic activity helps to capture the dynamics demarcated by the Gaussian activation pertaining to the input space. This practice leads to a remarkable reduction in curse of dimensionality. The proposed NN architecture has been compared with two other conventional architectures; one with dynamic neural units (DNUs) and the other with nonlinear static functions as perceptrons. The objective is to control evaporator heat flow rate and secondary fluid outlet temperature while keeping the degree of refrigerant superheat in the range 4–7 K at the evaporator outlet by manipulating refrigerant and evaporator secondary fluid flow rates. The drawbacks of conventional approaches to this problem are discussed, and how the novel method can overcome them are presented. An evolutionary approach is adopted to optimize the parameters of the NN controllers. Then evaporator process model is accomplished as a combination of governing equations and a sub NN resulting in a simple and sufficiently accurate model. The effectiveness of the proposed dynamic NN controller for the evaporator system model is validated using experimental data from the ammonia refrigeration plant.     

Impairment of long-term associative memory in aging snails (Lymnaea stagnalis).

Age-dependent impairment in learning and memory functions occurs in many animal species, including humans. Although cell death contributes to age-related cognitive impairment in pathological forms of aging, learning and memory deficiencies develop with age even without substantial cell death. The molecular and cellular basis of this biological aging process is not well understood but seems to involve a decline in the aging brain's capacity for experience-dependent plasticity. To aid in resolving this issue, we used a simple snail appetitive classical conditioning paradigm in which the underlying molecular, cellular, and neural network functions can be directly linked to age-associated learning and memory performance (i.e., the Lymnaea stagnalis feeding system). Our results indicate that age does not affect the acquisition of appetitive memory but that retention and/or consolidation of long-term memory become progressively impaired with advancing age. The latter phenomenon correlates with declining electrophysiological excitability in key neurons controlling the feeding behavior. Together, these results present the Lymnaea feeding system as a powerful paradigm for investigations of cellular and molecular foundations of biological aging in the brain. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Local Protein Translation and RNA Processing of Synaptic Proteins in Autism Spectrum Disorder

Autism spectrum disorder (ASD) is a heritable neurodevelopmental condition associated with impairments in social interaction, communication and repetitive behaviors. While the underlying disease mechanisms remain to be fully elucidated, dysfunction of neuronal plasticity and local translation control have emerged as key points of interest. Translation of mRNAs for critical synaptic proteins are negatively regulated by Fragile X mental retardation protein (FMRP), which is lost in the most common single-gene disorder associated with ASD. Numerous studies have shown that mRNA transport, RNA metabolism, and translation of synaptic proteins are important for neuronal health, synaptic plasticity, and learning and memory. Accordingly, dysfunction of these mechanisms may contribute to the abnormal brain function observed in individuals with autism spectrum disorder (ASD). In this review, we summarize recent studies about local translation and mRNA processing of synaptic proteins and discuss how perturbations of these processes may be related to the pathophysiology of ASD.     

Therapeutic Effects of hiPSC-Derived Glial and Neuronal Progenitor Cells-Conditioned Medium in Experimental Ischemic Stroke in Rats

Transplantation of various types of stem cells as a possible therapy for stroke has been tested for years, and the results are promising. Recent investigations have shown that the administration of the conditioned media obtained after stem cell cultivation can also be effective in the therapy of the central nervous system pathology (hypothesis of their paracrine action). The aim of this study was to evaluate the therapeutic effects of the conditioned medium of hiPSC-derived glial and neuronal progenitor cells in the rat middle cerebral artery occlusion model of the ischemic stroke. Secretory activity of the cultured neuronal and glial progenitor cells was evaluated by proteomic and immunosorbent-based approaches. Therapeutic effects were assessed by overall survival, neurologic deficit and infarct volume dynamics, as well as by the end-point values of the apoptosis- and inflammation-related gene expression levels, the extent of microglia/macrophage infiltration and the numbers of formed blood vessels in the affected area of the brain. As a result, 31% of the protein species discovered in glial progenitor cells-conditioned medium and 45% in neuronal progenitor cells-conditioned medium were cell type specific. The glial progenitor cell-conditioned media showed a higher content of neurotrophins (BDNF, GDNF, CNTF and NGF). We showed that intra-arterial administration of glial progenitor cells-conditioned medium promoted a faster decrease in neurological deficit compared to the control group, reduced microglia/macrophage infiltration, reduced expression of pro-apoptotic gene Bax and pro-inflammatory cytokine gene Tnf, increased expression of anti-inflammatory cytokine genes (Il4, Il10, Il13) and promoted the formation of blood vessels within the damaged area. None of these effects were exerted by the neuronal progenitor cell-conditioned media. The results indicate pronounced cytoprotective, anti-inflammatory and angiogenic properties of soluble factors secreted by glial progenitor cells.     

多个神经振子群网络的相位动力学编码

利用随机相变动力学理论研究运动认知的神经网络动力学模型．给出了感觉神经元集群、中间神经元集群和运动皮层神经元集群在耦合条件下相互作用、相位编码和数密度随时间的演化．探讨了神经网络在自发运动条件下以及在刺激条件下的神经网络动力学响应．通过数值模拟证实了（1）Walter J．Freeman提出的皮层动力学响应不能够编码外刺激信息的猜想;（2）串行的神经网络系统的神经编码具有节律编码的性质;（3）在中枢神经系统的调控中,神经抑制有其重要的作用．     

随机信号在神经元网络中诱发的双空间相干共振

对确定性行为为静息的神经元网络施加随机信号进行控制,随着信号强度的增加,网络行为由无序到有序的空间行为一螺旋波再到无序,螺旋波的结构由复杂到简单再到复杂到简单的交替,由网络行为的空间结构函数计算出的信噪比会两次达到极大值,即发生了两次空间相干共振,结果不仅展示了该随机信号控制下的网络的动力学行为,还为通过施加控制因素诱导产生空间共振来提高神经系统的信息处理能力提供了可能的方法．     

异质性和时滞作用下神经元网络的共振动力学

利用参数互异的Fitzhugh-Nagumo神经元构建了含耦合时滞的无标度神经元网络模型,通过数值模拟的方法,提出研究参数异质性和耦合时滞影响下神经元网络的共振动力学.结果发现,当耦合项中不含时滞时,适中的参数异质性能够使得神经元网络对外界弱周期信号的响应达到最优,即适中的参数异质性能够诱导神经元网络的共振响应,而且异质性诱导共振对耦合强度具有鲁棒性.更重要的是,耦合时滞对参数异质性作用下神经元网络的共振特性也有着显著性影响.当时滞约为信号周期的整数倍时,神经元网络能够周期性地出现共振现象,即适当的耦合时滞能够诱导神经元网络的多重共振,而且这种现象在异质性参数的适当范围内都能明显出现.     

Expression of high-molecular-mass neurofilament protein in horse (Equus caballus) spinal ganglion neurons

Spinal ganglion (SG) neurons are subdivided, on the basis of their cytoplasmic aspect at light and electron microscopy, into dark (D) and light (L) neurons. Numerous efforts have been made to find specific markers able to identify D and L neuronal cytotypes. The isolectin B4 (IB4), utilized to identify nonpeptidergic D neurons in mice, unfortunately, has not proved as effective in other species. The 200-kDa neurofilament protein (NF200) is considered as a typical marker of L neurons in the rat, cat, and chick. The aim of this study was to analyze the histological, morphometric, and neurochemical characteristic of NF200-immunoreactive (IR) horse SG neurons, to better characterize them morphologically and functionally. NF200-IR neurons showed two levels (strong and weak) of staining intensity. Most (84%) strongly stained NF200-IR neurons corresponded to L neurons, and showed similar bimodality as in the size distribution study, which seems to indicate a third population of neurons, in addition to the two populations (small and large) previously identified. In triple-staining experiments where NF200 was colocalized with IB4, substance P (SP), and neuronal nitric oxide synthase (nNOS) neuronal markers, most NF200-IR neurons were single stained. On the contrary, most IB4-, SP-, and nNOS-stained neurons were triple labeled and almost equally subdivided between strong and weak NF200-IR with the latter being always smaller in size than strong NF200-IR neurons. In conclusion, horse SG neurons display significant morphometric and neurochemical differences compared with those of rodents.     

Differential Fos Expression Following Aspiration, Electrolytic, or Excitotoxic Lesions of the Perirhinal Cortex in Rats.

The authors explored the possibility that there are different neural consequences, beyond the primary site of brain damage, following perirhinal cortex (PRh) lesions made in different ways. Fos expression was used as a marker for neuronal activation and compared across the forebrains of rats that underwent the different types of surgery. Electrolytic and excitotoxic PRh lesions produced dramatic increases in Fos expression in the cortex, and excitotoxic and aspiration PRh lesions increased Fos expression in the dentate gyrus. These data are consistent with the hypothesis that different lesion methods have separable effects on neural function in regions outside the lesion site that could account for inconsistencies in the literature regarding the behavioral effects of PRh lesions on tests of spatial memory. (PsycINFO Database Record (c) 2010 APA, all rights reserved)