Architectural Parameters Research Articles

Sleep disorders in pediatric patients with agenesis of the corpus callosum.

There is limited information about sleep in agenesis of the corpus callosum (ACC). We aim to describe the sleep architecture and respiratory parameters of children with ACC. We performed a retrospective study of 20 patients with ACC who had polysomnography between 2000 and 2023. Demographic data, body mass index or weight for length, associated conditions, and polysomnography findings were collected. National Sleep Foundation sleep quality indicators as well as increased polysomnography arousal index ≥ 10 events/h were used in the analysis. Fisher's exact test or unpaired t test was used to compare groups. Average age was 5.9 ± 5.4 years old. A total of 12/20 patients were male; 6/20 were overweight/obese; 14/20 had complete ACC, and 6/20 had partial ACC; 8/20 had seizures; 15/20 had ≥ 1 National Sleep Foundation poor sleep quality indicator (decreased sleep efficiency [45%], decreased rapid eye movement sleep [53%]); and 9/20 had increased arousals. Between complete and partial ACC, there was no difference in presence of ≥ 1 poor sleep quality indicator (P = .61), sleep efficiency (P = .34), rapid eye movement sleep (P = .28), and arousals (P = 1.0). 11/18 had obstructive sleep apnea (OSA); 5/11 had associated central sleep apnea. There was no difference in OSA between those with complete and partial ACC (P = 1.0). OSA was associated with children < 3 years old (P = .01). Children with ACC have poor sleep quality, and many have OSA. There was no difference in sleep quality or presence of OSA between those with complete and partial ACC. OSA was seen more in younger children. Our study supports the need for screening of sleep-related disorders in patients with ACC. Kwon A, Gu PK, Zhang C, Davidson Ward SL, Perez IA. Sleep disorders in pediatric patients with agenesis of the corpus callosum. J Clin Sleep Med. 2024;20(10):1663-1667.

Design and optimization of staggered fin structure of heat exchanger based on Machine learning

A multi-scale coupling method based on the unit heat transfer model is proposed to simulate the flow heat transfer performance of a heat exchanger, in view of the complex structure of the heat transfer unit model inside the staggered tooth type heat exchanger and the characteristics of coolant flow. The numerical simulation results are compared with experimental results, and the errors of the simulation results under different working conditions are small, indicating that the modelling method is reliable. On this basis, the heat transfer unit model parameters of staggered fins were established by three-dimensional numerical simulations. The heat transfer performance of the heat exchanger under different structures was compared using a comprehensive evaluation factor, i.e., j/f. The results and experimental data show that the best flow heat transfer performance is achieved with a liquid-side fin length of 7.2 mm, an oil-side fin length of 4.8 mm, a liquid-side fin height of 3.1 mm and an oil-side fin height of 3.1 mm, an inlet flow velocity of 0.2 m/s and a flow angle of 90°. Later, an agent model was established using backward propagation (BP) artificial neural network (ANN) to predict the flow heat transfer performance of the heat transfer unit model, and it was concluded that the f-factor, j-factor and j/f predicted by the ANN model matched well with the simulated values, and the predicted trends of different indicators were the same as the real law, which proved that the error between the predicted and actual values was small, and the above structural parameters were the best architectural parameters, which eventually improved the heat transfer performance by 17 %.

The Use of Transparent Structures to Improve Light Comfort in Library Spaces and Minimize Energy Consumption: A Case Study of Warsaw, Poland

Light plays a key role in shaping the quality and atmosphere of interior spaces, and its importance and contradictions are amplified in the design of libraries. This study focuses on the problems associated with both insufficient natural light and excessive light. In both cases, visual discomfort is the result. The reason for these problems lies in the use of translucent structures with different parameters and properties in library architecture. This study analyzed the lighting environment in the main reading rooms of two university libraries in Warsaw. The research methods were based on a field survey of objects and an analysis of the architectural and construction parameters of the premises, as well as the physical and mechanical properties of various types of translucent materials used in the reading rooms. The results of this study shed light on the relationship between the interior space of the reading room and the geometry of transparent structures, as well as the effectiveness of daylighting in libraries in the natural conditions of Warsaw. The key point of the present study was to recognize the critical role that transparent materials and structures play in mitigating potential operational problems in library spaces. Wise selection of these elements at the design stage can help minimize problems related to thermal energy regulation, ventilation control, acoustic insulation, and increased visual comfort. The importance of this research area lies in its potential to optimize library buildings by increasing energy efficiency and reducing reliance on artificial lighting, heating, and air conditioning. Careful consideration of transparent materials at the design stage promises long-term benefits in the form of sustainable library spaces that not only meet functional requirements, but also contribute to a more environmentally conscious architectural landscape.

Biases in ARG-Based Inference of Historical Population Size in Populations Experiencing Selection.

Inferring the demographic history of populations provides fundamental insights into species dynamics and is essential for developing a null model to accurately study selective processes. However, background selection and selective sweeps can produce genomic signatures at linked sites that mimic or mask signals associated with historical population size change. While the theoretical biases introduced by the linked effects of selection have been well established, it is unclear whether ancestral recombination graph (ARG)-based approaches to demographic inference in typical empirical analyses are susceptible to misinference due to these effects. To address this, we developed highly realistic forward simulations of human and Drosophila melanogaster populations, including empirically estimated variability of gene density, mutation rates, recombination rates, purifying, and positive selection, across different historical demographic scenarios, to broadly assess the impact of selection on demographic inference using a genealogy-based approach. Our results indicate that the linked effects of selection minimally impact demographic inference for human populations, although it could cause misinference in populations with similar genome architecture and population parameters experiencing more frequent recurrent sweeps. We found that accurate demographic inference of D. melanogaster populations by ARG-based methods is compromised by the presence of pervasive background selection alone, leading to spurious inferences of recent population expansion, which may be further worsened by recurrent sweeps, depending on the proportion and strength of beneficial mutations. Caution and additional testing with species-specific simulations are needed when inferring population history with non-human populations using ARG-based approaches to avoid misinference due to the linked effects of selection.

Effect of solution pH on root architecture of four apple rootstocks grown in an aeroponics nutrient misting system.

The pH of the solution in the rhizosphere is an important factor that determines the availability and mobility of nutrients for plant uptake. Solution pH may also affect the root distribution and architecture of apple rootstocks. In this study, we evaluated the effect of solution pH on root system development of apple rootstocks using an aeroponics system designed and developed at Cornell AgriTech Geneva, USA. Four Geneva® apple rootstocks (G.210, G.214, G.41, and G.890) were grown in an aeroponic system under nutrient solution misting featuring continuously adjusted pH levels to three pH treatments (5.5, 6.5, and 8.0). Root development was monitored for 30 days and evaluated regularly for distribution and root mass. Images of the developed roots grown in the aeroponic system were collected at the end of the experiment using a high-resolution camera and analyzed using GiA Roots® software, which generates root architecture parameter values in a semi-automated fashion. The resulting root architecture analysis showed that the Geneva® rootstocks were significantly different for two architecture parameters. The length-to-width ratio analysis represented by two GiA Roots parameters (minor-to-major ellipse ratio and network width-to-depth ratio) showed that G.210 was flatter than G.890, which had a greater tendency to grow downward. Rootstocks G.214 and G.41 displayed similar growth values. The solution pH affected most root architecture parameter measurements where overall root growth was higher at pH 8 than at pH 5.5 and 6.5, which showed similar growth. In general, the average root width tended to decrease at higher pH values. While there were no significant differences in the leaf nutrient concentrations of P, K, Ca, Mg, S, B, Zn, Cu, and Fe within the four rootstocks, the pH level of the solution had a significant effect on P, Ca, and Mn. This study is the first of its kind to investigate the effect of pH on root architecture in a soil-free (aeroponic) environment and may have implications for apple root behavior under field conditions where pH levels are different.

Method for Constructing Neural Network Means for Recognizing Scenes of Political Extremism in Graphic Materials of Online Social Networks

Countering the spread of calls for political extremism through graphic content on online social networks is becoming an increasingly pressing problem that requires the development of new technological solutions, since traditional approaches to countering are based on the results of recognizing destructive content only in text messages. Since in modern conditions neural network tools for analyzing graphic information are considered the most effective, it is assumed that it is advisable to use such tools for analyzing images and video materials in online social networks, taking into account the need to adapt them to the expected conditions of use, which are determined by the wide variability in the size of graphic content, the presence of typical interference, limited computing resources of recognition tools. Using this thesis, a method has been proposed that makes it possible to implement the construction of neural network recognition tools adapted to the specified conditions. For recognition, the author's neural network model was used, which, due to the reasonable determination of the architectural parameters of the low-resource convolutional neural network of the MobileNetV2 type and the recurrent neural network of the LSTM type, which makes up its structure, ensures high accuracy of recognition of scenes of political extremism both in static images and in video materials under limited computing conditions resources. A mechanism was used to adapt the input field of the neural network model to the variability of the size of graphic resources, which provides for scaling within acceptable limits of the input graphic resource and, if necessary, filling the input field with zeros. Levelling out typical noise is ensured by using advanced solutions in the method for correcting brightness, contrast and eliminating blur of local areas in images of online social networks. Neural network tools developed on the basis of the proposed method for recognizing scenes of political extremism in graphic materials of online social networks demonstrate recognition accuracy at the level of the most well-known neural network models, while ensuring a reduction in resource intensity by more than 10 times. This allows the use of less powerful equipment, increases the speed of content analysis, and also opens up prospects for the development of easily scalable recognition tools, which ultimately ensures an increase in security and a reduction in the spread of extremist content on online social networks. It is advisable to correlate the paths for further research with the introduction of the Attention mechanism into the neural network model used in the method, which will make it possible to increase the efficiency of neural network analysis of video materials.

Improving few-shot named entity recognition via Semantics induced Optimal Transport

Named entity recognition (NER) is to identify and categorize entities in unstructured text, which serves as a fundamental task for a variety of natural language processing (NLP) applications. In particular, emerging few-shot NER methods aim to learn model parameters well with few samples and have received considerable attention. The dominant few-shot NER methods usually employ pre-trained language models (PLMs) as their basic architecture and fine-tune model parameters with few NER samples. Since the sample size is small and there are a large number of parameters in PLMs, fine-tuning may result in the parameters of PLMs being highly biased. To address this issue, this study introduces the semantic distribution distance constraints to optimize the fine-tuning process of few-shot NER models and develops a framework named Semantic Constraints on few-shot Named Entity Recognition (SCNER). Specifically, the framework formulates the general knowledge transfer of PLMs as an optimal transport procedure with a semantic prior. And, a Semantics-induced Optimal Transport (SOT) regularizer is developed to utilize the importance and similarities of tokens within sentences. SOT builds the semantic distribution of the sentence and defines the transport costs between tokens to achieve the token-level optimal transport procedures. Finally, SOT is employed as a regularization term of few-shot NER to introduce the semantic distribution distance constraint for effectively transferring general knowledge from PLMs. The experiments on four public datasets demonstrate that the proposed method significantly improves the performance of NER models in both few-shot and fully supervised scenarios. SCNER is a common framework that can be applied to a variety of models without adding additional learning parameters, and can be used to enhance the generalization ability and adaptability of various few-shot NER models.

An explainable & prescriptive solution for space-based energy consumption optimization using BIM data & genetic algorithm

Creating energy-efficient buildings is a multifaceted challenge that involves carefully considering architectural, mechanical, and electrical parameters. With the increasing emphasis on low-energy building mandates and sustainable construction practices, integrating energy performance simulation into the design process has become imperative. However, existing methods often struggle to handle the vast amount of design information and to explore optimal alternatives effectively. While recent approaches such as Building Information Models (BIM) and data-driven solutions show promise for energy analysis, they frequently lack transparency and robust design optimization capabilities, limiting their applicability in critical contexts within the Architecture, Engineering, Construction, and Operation (AECO) industry. In response to these challenges, this research proposes a novel prescriptive model aimed at addressing these limitations. The proposed methodology integrates building energy simulation with optimization techniques, utilizing BIM data and a Genetic Algorithm (GA). The Genetic algorithm provides some optimized solutions, including different alternatives of geometry, material, electrical, and mechanical elements to help engineering for data-driven decision-making. This integration aims to develop a model that is both prescriptive and explainable for indoor building design. Leveraging the value engineering method, the proposed approach seeks to strike a balance between energy consumption, functionality, and cost. By doing so, it aims to enhance energy efficiency while providing tangible design optimization solutions.

How Close to the Mark Might Published Heritability Estimates Be?

The behavioural scientist who requires an estimate of narrow heritability, h2, will conduct a twin study, and input the resulting estimated covariance matrices into a particular mode of estimation, the latter derived under supposition of the standard biometric model (SBM). It is known that the standard biometric model can be expected to misrepresent the phenotypic (genetic) architecture of human traits. The impact of this misrepresentation on the accuracy of h2 estimation is unknown. We aimed to shed some light on this general issue, by undertaking three simulation studies. In each, we investigated the parameter recovery performance of five modes- Falconer’s coefficient and the SEM models, ACDE, ADE, ACE, and AE- when they encountered a constructed, non-SBM, architecture, under a particular informational input. In study 1, the architecture was single-locus with dominance effects and genetic-environment covariance, and the input was a set of population covariance matrices yielded under the four twin designs, monozygotic-reared together, monozygotic-reared apart, dizygotic-reared together, and dizygotic-reared apart; in study 2, the architecture was identical to that of study 1, but the informational input was monozygotic-reared together and dizygotic-reared together; and in study 3, the architecture was multi-locus with dominance effects, genetic-environment covariance, and epistatic interactions. The informational input was the same as in study 1. The results suggest that conclusions regarding the coverage of h2 must be drawn conditional on a) the general class of generating architecture in play; b) specifics of the architecture’s parametric instantiations; c) the informational input into a mode of estimation; and d) the particular mode of estimationemployed. The results showed that the more complicated the generating architecture, the poorer a mode’s h2 recovery performance. Random forest analyses furthermore revealed that, depending on the genetic architecture, h2, the dominance and locus additive parameter, and proportions of alleles were involved in complex interaction effects impacting on h2 parameter recovery performance of a mode of estimation. Data and materials: https://osf.io/aq9sx/

Comparative Analysis of the Performance of Artificial Neural Networks in Assessing the Technical Condition of Steel Ropes

Introduction. Currently, artificial neural networks (ANN) are successfully used for technical diagnostics of steel ropes. Expensive software products with an adapted neural network implementation environment, such as STATISTICA, Amygdala, MatLab Simulink, are often used for this purpose. The most affordable way to build and train an ANN, from a financial point of view, is to write your own program code using interactive libraries such as TensorFlow, PyTorch, Scikit-learn. However, such libraries are not fully adapted for building an ANN, and to use them you need to have basic programming skills. As a result, the quality of an ANN depends not only on its architecture, training data, and composition, but also on the environment in which it is built. The aim of the work was to compare the quality of the ANN, built and trained by various methods according to the criterion of test network performance, confidence levels for assessing the technical condition of the rope, as well as the complexity and speed of training. For this purpose, new software has been developed to solve the problem of assessing the technical condition of a steel rope using a combination of various rejection indicators. Materials and Methods. The basis for an ANN training was a statistical database of typical damages of steel ropes and, an expert assessment of the technical condition of steel ropes. The software was written in the Python programming language. Various methods of programming a neural network were presented: an ANN built on the basis of the STATISTICA software package and an ANN built using the interactive Scikit-learn library. Ten test samples were prepared to verify the operation of the ANN. The ANN quality was assessed based on the test network performance and confidence probabilities (activation levels of the “winning” neuron) of determining the technical condition of the rope. Results. The construction of the ANN using the interactive library Scikit-learn showed a relatively high complexity of construction and a relatively low learning rate of the ANN. Test performance of the network, with a test sample size of ten, turned out to be the same for both built ANNs. At the same time, there was a difference in the indicator of the average confidence level for determining the technical condition of a steel rope between the results of the ANN built on the basis of the STATISTICA software package and the ANN built using the Scikit-learn interactive library. Discussion and Conclusion. The results showed that the ANN built using the STATISTICA software package with the same architecture and network learning parameters had more optimal software algorithms according to the criteria of confidence probability and network learning speed in comparison with the ANN built using the free Skicit-learn library. However, the indicator of the ANN test performance turned out to be the same for both ANNs. This result justified the use of TensorFlow, PyTorch, and Skicit-learn libraries by the world's leading research and commercial centers in the field of artificial intelligence. The obtained scientific result allows us to numerically evaluate and compare the quality of an ANN having the same architecture and learning parameters, but built using different methods. This will be useful for future scientific research in the field and for selecting the optimal environment for constructing ANNs in industrial applications.

Design of Software Reliability Prediction using Radial Basis Function Networks with Nonlinear Analysis and Topological Considerations

This study presents a thorough methodology for predicting software reliability by employing Radial Basis Function Networks (RBFNs). By using machine learning methodologies such as Radial Basis Function Networks (RBFNs), software development teams are enabled to make well-informed decisions, optimize resource allocation, and proactively mitigate potential dependability concerns. Consequently, this results in improved software quality and heightened customer happiness. This methodology involves multiple stages, starting with data collection and preprocessing. We assemble a comprehensive dataset comprising historical software performance metrics, defect reports, and relevant development process information. After data cleansing and feature engineering, we split the dataset into training, validation, and testing subsets. The core of this approach lies in the construction and training of RBFNs. These neural networks consist of an input layer, a hidden layer with radial basis functions, and an output layer. The architecture parameters, particularly the number of hidden neurons and the spread parameter of the radial basis functions, are optimized through a systematic hyperparameter tuning process using the validation dataset. Upon achieving an optimal model configuration, we rigorously evaluate the RBFN predictive capabilities using the testing dataset.

Canopy Architectural Characteristics of Ten New Olive (Olea europaea L.) Genotypes and Their Potential for Cultivation in Super-High-Density Orchards.

In recent years, there has been growing interest in olive genotypes (Olea europaea L.) suitable for super-high-density (SHD > 1200 trees/hectare) orchards. To date, only a few cultivars are considered fitting for such cultivation system. In this study, the first results on the architectural characteristics of the canopy of ten new olive genotypes are presented. Their suitability for SHD orchards was evaluated and compared with the cultivar 'Arbequina', which is considered suitable for SHD olive orchards and, for this reason, was used as the control. Several canopy measurements were taken, and some architectural parameters, such as branching frequency, branching density, and branch diameter/stem diameter ratio were calculated. The branching frequency value was greater than 0.20 in 'Arbequina' and in only four of the genotypes. The branching density in five genotypes was similar to 'Arbequina'. 'Arbequina' had the lowest value for the branch diameter/stem diameter ratio, and only three genotypes had similar values. These initial results showed that only one genotype has all canopy architectural characteristics comparable to those of the cv. 'Arbequina'. Further studies are needed to evaluate the production traits of these new genotypes and complete their characterization.

Precision Phenotyping of Wild Rocket (Diplotaxis tenuifolia) to Determine Morpho-Physiological Responses under Increasing Drought Stress Levels Using the PlantEye Multispectral 3D System

The PlantEye multispectral scanner is an optoelectrical sensor automatically applied to a mechatronic platform that allows the non-destructive, accurate, and high-throughput detection of morphological and physiological plant parameters. In this study, we describe how the advanced phenotyping platform precisely assesses changes in plant architecture and growth parameters of wild rocket salad (Diplotaxis tenuifolia L. [DC.]) under drought stress conditions. Four different irrigation supply levels from moderate to severe, required to keep 100, 70, 50, and 30% of the water-holding capacity, were adopted. Growth rate and plant architecture were recorded through the digital measure of biomass, leaf area, Canopy Light Penetration Depth, five convex hull traits, plant height, Surface Angle Average, and Voxel Volume Total. Vegetation color assessments included hue, lightness, and saturation. Vegetation and senescence indices were calculated from canopy reflectance in the red (620–645 nm), green (530–540 nm), blue (peak wavelength 460–485 nm), near-infrared (820–850 nm), and 3D laser (940 nm) ranges. The temperature, relative humidity, and solar radiation of the environment were also recorded. Overall, morphological parameters, color, multispectral data, and vegetation indices provided over 7200 data points through daily scans over three weeks of cultivation. Although a general decrease in growth parameters with increasing stress severity was observed, plants were able to maintain the same morpho-physiological performances as the control during the early growth stages, keeping both 70% and 50% of the total water-holding capacity. Among indices, the Normalized Differential Vegetation Index (NDVI) contributed the most to the differentiation between different stress levels during the cultivation cycle. Across the 3 weeks of growth, statistically significant differences were observed for all traits except for the Saturation Average. Comparisons with respect to the control highlighted the strong impact of drought stress on morphological plant traits. This study provided meaningful insights into the health status of wild rocket salad under increasing drought stress.

Decoupled differentiable graph neural architecture search

The differentiable graph neural architecture search (GNAS) effectively designs graph neural networks (GNNs) efficiently and automatically with excellent performance based on different graph data distributions. Given a GNN search space containing multiple GNN component operation candidates, the differentiable GNAS method builds a mixed supernet using learnable architecture parameters multiplied by the GNN component operation candidates. When the mixed supernet completes optimization, the mixed supernet is pruned based on the best architecture parameters to efficiently identify the optimal GNN architecture in the GNN search space. However, the multiplicative relationship between the architecture parameters and the GNN component operation candidates introduces a coupled optimization bias into the weight optimization process of the mixed supernet GNN component operation candidates. This bias results in differentiable GNAS performance degradation. To solve the problem of coupled optimization bias in the previous differentiable GNAS method, we propose the Decoupled Differentiable Graph Neural Architecture Search (D2GNAS). It utilizes the Gumbel distribution as a bridge to decouple the weights optimization of supernet GNN component candidate operation and architecture parameters for constructing the decoupled differentiable GNN architecture sampler. The sampler is capable of selecting promising GNN architectures based on architecture parameters treated as sampling probabilities, and it is further optimized through the validation gradients derived from the sampled GNN architectures. Simultaneously, D2GNAS builds a single-path supernet with a pruning strategy to compress the supernet progressively to improve search efficiency further. We conduct extensive experiments on multiple benchmark graphs. The experimental findings demonstrate that D2GNAS outperforms all established baseline methods, both manual GNN and GNAS methods, in terms of performance. Additionally, D2GNAS has a lower time complexity than previous differentiable GNAS methods. Based on the fair GNN search space, it achieves an average 5x efficiency improvement. Codes are available at https://github.com/AutoMachine0/D2GNAS.

Reproducibility of freehand vs. foam cast as well as the intrarater reliability of foam cast ultrasound scans assessing the muscle architecture and tissue organization of the gastrocnemius medialis and vastus lateralis muscles.

This study compares the reproducibility of freehand (FH) vs. foam cast (FC) scans and investigates the intrarater reliability of the ultrasound FC muscle architecture and tissue organization measurements of the gastrocnemius medialis (GM) and vastus lateralis (VL) muscles with fixed and repositioning FC scans. Thirteen young adults (22 ± 3 years) underwent repeated sagittal B-mode ultrasound measurements of GM and VL. FH, FC, and repositioned FC scans were conducted. Muscle architecture measurements included muscle thickness (MT), pennation angle (PA), and fascicle length (FL). Spatial frequency analysis assessed muscle tissue organization. MT decreased from 2.1 to 1.8 cm in GM and from 2.4 to 2.2 cm in VL with the FC compared with the FH. Reproducibility between the FH and the FC showed poor to good intraclass correlation coefficients (ICCs) for MT (0.46-0.77) and PA (0.09-0.86) as well as poor to moderate ICCs for FL (0.41), with very low to moderate test-retest variability (TRV) (4%-18%). Tissue organization indicated low to good ICCs (0.21-0.80) with low to moderate TRV (4%-19.5%). The re-scanning results of fixed FC indicated excellent ICCs for MT (0.95-0.996), good for PA (0.77-0.90), and moderate for FL (0.73-0.76), with low TRV (5%-10%) for both muscles. Tissue organization displayed moderate to good ICCs (0.61-0.87) with very low to low TRV (4%-9%). For repositioned FC scans in GM and VL, MT showed good to excellent ICCs (0.86-0.98) with very low to low TRV (2%-8%). PA and FL demonstrated moderate to good ICCs (0.57-0.75), with very low to moderate TRV (2%-13%). Tissue organization revealed ICCs ranging from poor to good (0.13-0.87) for both muscles, with low to moderate TRV (5%-18%). The FC systematically reduced MT by 2-3 mm. Furthermore, reproducibility revealed low ICCs and high data variability for several muscle architecture and tissue organization parameters. Thus, switching methods within a single study is not recommended. Nevertheless, FC ultrasound scans demonstrated excellent intrarater reliability for assessing MT. In the case of fixed FC scans particularly, moderate to excellent ICCs were observed for all muscle architecture and tissue organization parameters, accompanied by very low to low variability. Therefore, FC scans are recommended for investigating acute effects on muscle architecture and tissue organization when the FC remains on the leg throughout the period of measurements.

Stochastic Integrated Actor-Critic for Deep Reinforcement Learning.

We propose a deep stochastic actor-critic algorithm with an integrated network architecture and fewer parameters. We address stabilization of the learning procedure via an adaptive objective to the critic's loss and a smaller learning rate for the shared parameters between the actor and the critic. Moreover, we propose a mixed on-off policy exploration strategy to speed up learning. Experiments illustrate that our algorithm reduces the sample complexity by 50%-93% compared with the state-of-the-art deep reinforcement learning (RL) algorithms twin delayed deep deterministic policy gradient (TD3), soft actor-critic (SAC), proximal policy optimization (PPO), advantage actor-critic (A2C), and interpolated policy gradient (IPG) over continuous control tasks LunarLander, BipedalWalker, BipedalWalkerHardCore, Ant, and Minitaur in the OpenAI Gym.

SpikeExplorer: Hardware-Oriented Design Space Exploration for Spiking Neural Networks on FPGA

One of today’s main concerns is to bring artificial intelligence capabilities to embedded systems for edge applications. The hardware resources and power consumption required by state-of-the-art models are incompatible with the constrained environments observed in edge systems, such as IoT nodes and wearable devices. Spiking Neural Networks (SNNs) can represent a solution in this sense: inspired by neuroscience, they reach unparalleled power and resource efficiency when run on dedicated hardware accelerators. However, when designing such accelerators, the amount of choices that can be taken is huge. This paper presents SpikExplorer, a modular and flexible Python tool for hardware-oriented Automatic Design Space Exploration to automate the configuration of FPGA accelerators for SNNs. SpikExplorer enables hardware-centric multiobjective optimization, supporting target factors such as accuracy, area, latency, power, and various combinations during the exploration process. The tool searches the optimal network architecture, neuron model, and internal and training parameters leveraging Bayesian optimization, trying to reach the desired constraints imposed by the user. It allows for a straightforward network configuration, providing the full set of explored points for the user to pick the trade-off that best fits their needs. The potential of SpikExplorer is showcased using three benchmark datasets. It reaches 95.8% accuracy on the MNIST dataset, with a power consumption of 180 mW/image and a latency of 0.12 ms/image, making it a powerful tool for automatically optimizing SNNs.

Three-dimensional architecture and moment arms of human rotator cuff muscles invivo: Interindividual, intermuscular, and intramuscular variations.

The human rotator cuff consists of four muscles, each with a complex, multipennate architecture. Despite the functional and clinical importance, the architecture of the human rotator cuff has yet to be clearly described in humans invivo. The purpose of this study was to investigate the intramuscular, intermuscular, and interindividual variations in architecture and moment arms of the human rotator cuff. Muscle volumes, fascicle lengths, physiological cross-sectional areas (PCSAs), pennation angles, and moment arms of all four rotator cuff muscles were measured from mDixon and diffusion tensor imaging (DTI) scans of the right shoulders of 20 young adults. In accordance with the most detailed dissections available to date, we found substantial intramuscular variation in fascicle length (coefficients of variation (CVs) ranged from 26% to 40%) and pennation angles (CVs ranged from 56% to 62%) in all rotator cuff muscles. We also found substantial intermuscular and interindividual variations in muscle volumes, but relatively consistent mean fascicle lengths, pennation angles, and moment arms (CVs for all ≤17%). Moreover, when expressed as a proportion of total rotator cuff muscle volume, the volumes of individual rotator cuff muscles were highly consistent between individuals and sexes (CVs ≤16%), suggesting that rotator cuff muscle volumes scale uniformly, at least in a younger population without musculoskeletal problems. Together, these data indicate limited interindividual and intermuscular variability in architecture, which may simplify scaling routines for musculoskeletal models. However, the substantial intramuscular variation in architecture questions the validity of previously reported mean architectural parameters to adequately describe rotator cuff function.

Defining Architectural Changes in Mouse Pelvic Floor Muscles During Pregnancy to Withstand Physiological Distension Associated with Parturition

Pelvic floor disorders (PFDs), which are present in ~25% of women, profoundly impact quality of life. During childbirth, pelvic floor muscles (PFMs) endure substantial mechanical strain, which can lead to PFM dysfunction. As such, vaginal childbirth is a major risk factor for PFDs. Previous studies in rats demonstrate important structural adaptations in the PFMs that help to withstand the mechanical strain of childbirth, including muscle fiber lengthening by sarcomerogenesis. However, it is diffcult to establish a comprehensive understanding of the molecular mechanisms underlying PFM dysfunction during childbirth in rat models. Here, our objective was to determine if mouse PFMs undergo similar adaptations to rats during pregnancy and vaginal distension. We hypothesized that, akin to rats, PFMs in mice would undergo muscle fiber elongation through sarcomerogenesis during pregnancy. The PFMs (coccygeus [C], iliocaudalis [ICa], pubocaudalis [PCa]) and a non-pelvic floor muscle (tibialis anterior [TA]) were collected from late-pregnant (E16.5) and non-pregnant C57BL/6NJ (3 months old) mice. These animals were subjected or not subjected to physiological vaginal distension to mimic vaginal parturition (N = 4/group); in anesthetized mice, a 6F transurethral catheter was inserted into the vagina, and it was inflated with 0.3 mL (which best approximated fetal head circumference), and a 13 g weight was attached to the catheter creating circumferential and downward strains similar to parturition. Following this, the animals were euthanized, and the pelvis was fixed in situ for assessment of muscle architectural parameters; muscle fibers length was determined using digital calipers and sarcomere length was measured by light microscopy. Pregnancy induced a significant increase in normalized muscle fiber length in all PFMs compared to non-pregnant animals (Pregnant vs. not pregnant — C: 4.71±0.06 mm vs. 3.37±0.15 mm, p&lt;0.0001; ICa 8.74±0.14 mm vs. 7.50±0.27 mm, p=0.004; PCa 7.13±0.09 mm vs. 6.28±0.34 mm, p=0.04); TA muscle fiber length was unchanged (6.92±0.17 mm vs. 6.73 ± 0.19 mm, p=0.28). Sarcomere length was not affected by pregnancy in any muscle (C: 2.44±0.002 μm vs. 2.45 ± 0.003 μm, p=0.55; ICa: 2.26±0.003 μm vs. 2.26±0.004 μm, p=0.64; PCa: 2.16±0.01 μm vs. 2.16±0.003 μm, p=0.74; TA: 2.55±0.004 μm vs. 2.54±0.003 μm, p=0.28). Balloon-mediated vaginal distention resulted in significantly longer sarcomere lengths in non-pregnant compared to pregnant animals in all PFMs (C: 2.98±0.01 μm vs. 2.43±0.01 μm, p&lt;0.0001; ICa: 2.54±0.004 μm vs. 2.24±0.005 μm, p&lt;0.0001; PCa: 2.85±0.001 μm vs. 2.17±0.001 μm, p&lt;0.0001), indicating protection against sarcomere elongation in pelvic floor muscles of pregnant animals. This study shows that, similar to rats, pregnancy causes architectural changes in mouse pelvic floor muscles, which prevent excessive sarcomere stretching during physiologic vaginal distension that is comparable to fetal delivery. This work was funded by NIH grant #K12 HD000849 (Reproductive Scientist Development Program) to LAB and UC San Diego Senate Grant to LAB. This is the full abstract presented at the American Physiology Summit 2024 meeting and is only available in HTML format. There are no additional versions or additional content available for this abstract. Physiology was not involved in the peer review process.

Hindlimb myology in two piscivorous raptorial birds: a quantitative comparison of the osprey and the white‐tailed sea eagle (Aves, Accipitriformes)

AbstractThe hindlimbs of Accipitriformes are vital for capturing prey. Research on hindlimb musculature has primarily focused on species preferring avian and mammalian prey, whereas piscivorous species have received limited attention. This study aims to investigate the quantitative characteristics of hindlimb muscles in two piscivorous Accipitriformes, Pandion haliaetus and Haliaeetus albicilla, to discern potential muscular features associated with their specific food preference. The mass and proportion of all hindlimb muscles in both species were assessed based on their primary function (flexion or extension). A Kruskal‐Wallis test was employed to analyze possible differences in muscle mass between species. The allometric relationships between the muscles and body mass were explored with the reduced major axis method. Additionally, a study on the architectural parameters of the primary gripping muscles in P. haliaetus was conducted, using published information from other raptorial birds for comparison. The isometric scaling relationship predominated in the majority of individual muscles helping maintain a proportional relationship relative to body mass. Both species exhibited a similar pattern in terms of quantitative muscle features, implying a preservation of muscle characteristics linked to their predatory capabilities. The largest proportion of hindlimb muscle mass was dedicated to digit flexion in accordance with the grasping abilities of birds of prey. The muscles tibialis cranialis, flexor digitorum longus and flexor hallucis longus exhibited the greatest mass, high physiological cross‐sectional area values, and long fibre, categorizing them as high‐power specialists. These muscles are crucial in the flexion movements that produce the gripping action that characterizes birds of prey. Although no statistically significant differences were detected, each species displayed slightly distinct muscular characteristics, particularly in the architectural properties of the flexor muscles controlling digits II, III, and IV. These variations seem to be associated with differences in their prey preferences.