Search

This paper proposes a new load-balancing method that supports Per-Connection Consistency (PCC) by leveraging ECMP with standardized routing protocols for operational simplicity. Load balancers must consistently deliver packets of a flow to one of the backend servers, even when the backend server pool changes. Achieving this property—PCC—is not a straightforward task and usually introduces additional complexity into networks, such as special packet forwarding mechanisms, synchronizing states across load balancers, and often complicated orchestration. Our approach, VRF-shadowing, achieves PCC on ECMP of routers without additional control plane systems by exploiting Segment Routing over IPv6 (SRv6). VRF-shadowing employs the daisy-chaining concept, achieving PCC in a stateless manner with a standardized SRv6-based Layer-3 VPN (L3VPN). VRF-shadowing is composed of two components: a minimal new feature for software of SRv6 routers and a new SRv6 End behavior at servers. We implemented the former in a commercial router and the latter in Linux and evaluated VRF-shadowing on an SRv6 L3VPN network built with only standardized routing protocols. The evaluation results demonstrate that VRF-shadowing preserves connections from being disrupted when the ECMP next-hops change. In the most severe scenario, VRF-shadowing reduced the failed HTTP requests rate from 77% to 0.7% compared to traditional ECMP.

Many countries have gradually shifted to animal welfare-friendly housing systems for egg layers. However, there is still no consensus among researchers on whether the housing system affects egg quality traits. Therefore, this study aimed to determine the effects of housing systems on egg traits and free amino acid contents of albumen and yolk using two types of housing systems, the conventional cage (cage) system and a floor rearing (litter) system. Tosa-jidori (n = 20) hens were divided into two groups. Experiments during the 7 weeks were performed twice by switching the housing systems (first and second stages). One-way analysis of variance was used to evaluate the effects of housing systems on body weight gain, egg traits, albumen and yolk amino acid contents, and fecal corticosterone. We observed significant housing effects in body weight gain, eggshell weight, yolk weight, eggshell thickness, eggshell lightness, and several albumen amino acids (A_Gln, A_His, A_Met, A_Cys, A_Lys, A_Asp, A_Glu, A_Ser, A_Thr, A_Ala, A_Pro, and A_Phe). Notably, a robust effect was seen in eggshell lightness, even after switching housing systems. These results suggest that eggshell lightness and several egg traits, including albumen amino acid contents, can be changed by using the different housing systems.

Electromyogram (EMG) classification is a key technique in EMG-based control systems. Existing EMG classification methods, which do not consider EMG features that have distribution with skewness and kurtosis, have limitations such as the requirement to tune hyperparameters. In this paper, we propose a neural network based on the Johnson $S_{\mathrm {U}}$ translation system that is capable of representing distributions with skewness and kurtosis. The Johnson system is a normalizing translation that transforms non-normal distribution data into normal distribution data, thereby enabling the representation of a wide range of distributions. In this study, a discriminative model based on the multivariate Johnson $S_{\mathrm {U}}$ translation system is transformed into a linear combination of coefficients and input vectors using log-linearization; then, it is incorporated into a neural network structure. This allows the calculation of the posterior probability of each class given the input vectors and the determination of model parameters as weight coefficients of the network. The uniqueness of convergence of the network learning is theoretically guaranteed. In the experiments, the suitability of the proposed network for distributions including skewness and kurtosis was evaluated using artificially generated data. Its applicability to real biological data was also evaluated via EMG classification experiments. The results showed that the proposed network achieved high classification performance (e.g., 99.973% accuracy using Khushaba’s dataset) without the need for hyperparameter optimization.

Convolutional Neural Networks (CNNs) achieve excellent computer-assisted diagnosis with sufficient annotated training data. However, most medical imaging datasets are small and fragmented. In this context, Generative Adversarial Networks (GANs) can synthesize realistic/diverse additional training images to fill the data lack in the real image distribution; researchers have improved classification by augmenting data with noise-to-image (e.g., random noise samples to diverse pathological images) or image-to-image GANs (e.g., a benign image to a malignant one). Yet, no research has reported results combining noise-to-image and image-to-image GANs for further performance boost. Therefore, to maximize the DA effect with the GAN combinations, we propose a two-step GAN-based DA that generates and refines brain Magnetic Resonance (MR) images with/without tumors separately: (i) Progressive Growing of GANs (PGGANs), multi-stage noise-to-image GAN for high-resolution MR image generation, first generates realistic/diverse 256×256 images; (ii) Multimodal UNsupervised Image-to-image Translation (MUNIT) that combines GANs/Variational AutoEncoders or SimGAN that uses a DA-focused GAN loss, further refines the texture/shape of the PGGAN-generated images similarly to the real ones. We thoroughly investigate CNN-based tumor classification results, also considering the influence of pre-training on ImageNet and discarding weird-looking GAN-generated images. The results show that, when combined with classic DA, our two-step GAN-based DA can significantly outperform the classic DA alone, in tumor detection (i.e., boosting sensitivity 93.67% to 97.48%) and also in other medical imaging tasks.

The effectiveness of biosignal generation and data augmentation with biosignal generative models based on generative adversarial networks (GANs), which are a type of deep learning technique, was demonstrated in our previous paper. GAN-based generative models only learn the projection between a random distribution as input data and the distribution of training data. Therefore, the relationship between input and generated data is unclear, and the characteristics of the data generated from this model cannot be controlled. This study proposes a method for generating time-series data based on GANs and explores their ability to generate biosignals with certain classes and characteristics. Moreover, in the proposed method, latent variables are analyzed using canonical correlation analysis (CCA) to represent the relationship between input and generated data as canonical loadings. Using these loadings, we can control the characteristics of the data generated by the proposed method. The influence of class labels on generated data is analyzed by feeding the data interpolated between two class labels into the generator of the proposed GANs. The CCA of the latent variables is shown to be an effective method of controlling the generated data characteristics. We are able to model the distribution of the time-series data without requiring domain-dependent knowledge using the proposed method. Furthermore, it is possible to control the characteristics of these data by analyzing the model trained using the proposed method. To the best of our knowledge, this work is the first to generate biosignals using GANs while controlling the characteristics of the generated data.

Although psychotropic drugs act on neurons and glial cells, how glia respond, and whether glial responses are involved in therapeutic effects are poorly understood. Here, we show that fluoxetine (FLX), an anti-depressant, mediates its anti-depressive effect by increasing the gliotransmission of ATP. FLX increased ATP exocytosis via vesicular nucleotide transporter (VNUT). FLX-induced anti-depressive behavior was decreased in astrocyte-selective VNUT-knockout mice or when VNUT was deleted in mice, but it was increased when astrocyte-selective VNUT was overexpressed in mice. This suggests that VNUT-dependent astrocytic ATP exocytosis has a critical role in the therapeutic effect of FLX. Released ATP and its metabolite adenosine act on P2Y11 and adenosine A2b receptors expressed by astrocytes, causing an increase in brain-derived neurotrophic factor in astrocytes. These findings suggest that in addition to neurons, FLX acts on astrocytes and mediates its therapeutic effects by increasing ATP gliotransmission. Keywords: Astrocytes, Fluoxetine, ATP, Adenosine, Vesicular nucleotide transporter, BDNF

In this study, we describe a novel mechanical vibration suppression method for high-speed contouring motions. In this method, vibration compensation torque was applied to suppress mechanical vibrations during high-speed contouring motions. The parameters of the compensation signal were determined using commanded acceleration and time for acceleration and deceleration, which were set as numerical control (NC) parameters. We also propose an autonomous compensation-torque-generation algorithm based on the acceleration command from NC. The effectiveness of this method was assessed by measuring and simulating rectangular corner contouring motions in which the compensation torque is applied. The results confirmed that the proposed method can effectively suppress the vibration when the motion direction changes. This method can also be effectively performed at various feed speeds by automatically adapting the torque based on the proposed criteria.

This paper proposes an artificial electromyogram (EMG) signal generation model based on signal-dependent noise, which has been ignored in existing methods, by introducing the stochastic construction of the EMG signals. In the proposed model, an EMG signal variance value is first generated from a probability distribution with a shape determined by a commanded muscle force and signal-dependent noise. Artificial EMG signals are then generated from the associated Gaussian distribution with a zero mean and the generated variance. This facilitates representation of artificial EMG signals with signal-dependent noise superimposed according to the muscle activation levels. The frequency characteristics of the EMG signals are also simulated via a shaping filter with parameters determined by an autoregressive model. An estimation method to determine EMG variance distribution using rectified and smoothed EMG signals, thereby allowing model parameter estimation with a small number of samples, is also incorporated in the proposed model. Moreover, the prediction of variance distribution with strong muscle contraction from EMG signals with low muscle contraction and related artificial EMG generation are also described. The results of experiments conducted, in which the reproduction capability of the proposed model was evaluated through comparison with measured EMG signals in terms of amplitude, frequency content, and EMG distribution demonstrate that the proposed model can reproduce the features of measured EMG signals. Further, utilizing the generated EMG signals as training data for a neural network resulted in the classification of upper limb motion with a higher precision than by learning from only measured EMG signals. This indicates that the proposed model is also applicable to motion classification.

BACKGROUNDAlthough the duration of untreated psychosis (DUP) plays an important role in the short-term prognosis of patients with schizophrenia, their long-term prognosis generally is not determined by DUP alone. It is important to explore how other clinical factors in the early stage are related to DUP and consequent disease courses.METHODSA total of 664 patients with untreated psychosis were surveyed for this study. At the first examination, we divided them into the severe positive symptoms cases (SC) or the less severe cases (NonSC) and compared the prognosis among the two groups after a 10-year follow-up. In all, 113 patients in the SC group and 43 patients in the NonSC group were follow-up completers.RESULTSWhereas DUP was not different between the two groups, patients with nonacute onset in both groups had significantly longer DUP than those in patients with acute onset. For all clinical measures, there was no difference in prognosis between the two groups or among the four groups classified by mode of onset (MoO) and initial severity of positive symptoms. However, the degree of improvement of global assessment of functioning (GAF) was significantly smaller in the NonSC-nonacute group than in the SC-acute and SC-nonacute groups.CONCLUSIONSThese results suggest that neither DUP nor MoO alone necessarily affects the initial severity of positive symptoms. Moreover, it is possible that patients with low impetus of positive symptoms onset within long DUP experience profound pathologic processes. Therefore, the current study results indicated that long DUP and nonacute onset were related to poor long-term prognosis, regardless of initial positive symptoms.

Polyglutamine (polyQ) expansion mutation causes conformational, neurodegenerative diseases, such as Alzheimer's and Parkinson's diseases. These diseases are characterized by the aggregation of misfolded proteins, such as amyloid fibrils, which are toxic to cells. Amyloid fibrils are formed by a nucleated growth polymerization reaction. Unexpectedly, the critical nucleus of polyQ aggregation was found to be a monomer, suggesting that the rate-limiting nucleation process of polyQ aggregation involves the folding of mutated protein monomers. The monoclonal antibody 1C2 selectively recognizes expanded pathogenic and aggregate-prone glutamine repeats in polyQ diseases, including Huntington's disease (HD), as well as binding to polyleucine. We have therefore assayed the in vitro and in vivo aggregation kinetics of these monomeric proteins. We found that the repeat-length-dependent differences in aggregation lag times of variable lengths of polyQ and polyleucine tracts were consistently related to the integration of the length-dependent intensity of anti-1C2 signal on soluble monomers of these proteins. Surprisingly, the correlation between the aggregation lag times of polyQ tracts and the intensity of anti-1C2 signal on soluble monomers of huntingtin precisely reflected the repeat-length dependent age-of-onset of HD patients. These data suggest that the alterations in protein surface structure due to polyQ expansion mutation in soluble monomers of the mutated proteins act as an amyloid-precursor epitope. This, in turn, leads to nucleation, a key process in protein aggregation, thereby determining HD onset. These findings provide new insight into the gain-of-function mechanisms of polyQ diseases, in which polyQ expansion leads to nucleation rather than having toxic effects on the cells.