Neural Network Model Research Articles

Distinct ventral hippocampal inhibitory microcircuits regulating anxiety and fear behaviors

In emotion research, anxiety and fear have always been interconnected, sharing overlapping brain structures and neural circuitry. Recent investigations, however, have unveiled parallel long-range projection pathways originating from the ventral hippocampus, shedding light on their distinct roles in anxiety and fear. Yet, the mechanisms governing the emergence of projection-specific activity patterns to mediate different negative emotions remain elusive. Here, we show a division of labor in local GABAergic inhibitory microcircuits of the ventral hippocampus, orchestrating the activity of subpopulations of pyramidal neurons to shape anxiety and fear behaviors in mice. These findings offer a comprehensive insight into how distinct inhibitory microcircuits are dynamically engaged to encode different emotional states.

A Comparative Study of Machine Learning and Conventional Techniques in Predicting Compressive Strength of Concrete with Eggshell and Glass Powder Additives

Recent research has focused on assessing the effectiveness of response surface methodology (RSM), a non-machine learning technique, and artificial neural networks (ANN), a machine learning approach, for predicting concrete performance. This research aims to predict and simulate the compressive strength of concrete that replaces cement and fine aggregate with waste materials such as eggshell powder (ESP) and waste glass powder (WGP) for sustainable construction materials. In order to ensure concrete’s durability and structural integrity, a compressive strength evaluation is essential. Precise predictions maximize efficiency and advance sustainability, particularly when dealing with waste materials like ESP and WGP. The response surface methodology (RSM) and artificial neural network (ANN) techniques are used to accomplish this for practical applications in the built environment. A dataset comprising previously published research was used to assess ANN and RSM’s predictive and generalization abilities. To model and improve the model, ANN used seven independent variables, while three variables, cement, waste glass powder, and eggshell powder, improved the RSM. Both the ANN and RSM techniques are effective instruments for predicting compressive strength, according to the statistical results, which include mean squared error (MSE), determination coefficient (R2), and adjusted coefficient (R2 adj). RSM was able to achieve the R2 by 0.8729 and 0.7532 for compressive strength, while the accuracy of the results for ANN was 0.907 and 0.956 for compressive strength. Moreover, the correlation between ANN and RSM models and experimental data is high. The artificial neural network model, however, exhibits superior accuracy.

A variant-informed decision support system for tackling COVID-19: a transfer learning and multi-attribute decision-making approach

The global impact of the COVID-19 pandemic, characterized by its extensive societal, economic, and environmental challenges, escalated with the emergence of variants of concern (VOCs) in 2020. Governments, grappling with the unpredictable evolution of VOCs, faced the need for agile decision support systems to safeguard nations effectively. This article introduces the Variant-Informed Decision Support System (VIDSS), designed to dynamically adapt to each variant of concern’s unique characteristics. Utilizing multi-attribute decision-making (MADM) techniques, VIDSS assesses a country’s performance by considering improvements relative to its past state and comparing it with others. The study incorporates transfer learning, leveraging insights from forecast models of previous VOCs to enhance predictions for future variants. This proactive approach harnesses historical data, contributing to more accurate forecasting amid evolving COVID-19 challenges. Results reveal that the VIDSS framework, through rigorous K-fold cross-validation, achieves robust predictive accuracy, with neural network models significantly benefiting from transfer learning. The proposed hybrid MADM approach integrated approaches yield insightful scores for each country, highlighting positive and negative criteria influencing COVID-19 spread. Additionally, feature importance, illustrated through SHAP plots, varies across variants, underscoring the evolving nature of the pandemic. Notably, vaccination rates, intensive care unit (ICU) patient numbers, and weekly hospital admissions consistently emerge as critical features, guiding effective pandemic responses. These findings demonstrate that leveraging past VOC data significantly improves future variant predictions, offering valuable insights for policymakers to optimize strategies and allocate resources effectively. VIDSS thus stands as a pivotal tool in navigating the complexities of COVID-19, providing dynamic, data-driven decision support in a continually evolving landscape.

Assistance of aminated straw on enhanced biodegradation of phenol through bacteria and potential prediction of machine learning for bioremediation

ABSTRACT How to treat phenol-containing wastewater harmlessly is an urgent problem to be solved. In this study, a kind of biomaterial was prepared through strain PAO1 immobilized on aminated straw to enhance the phenol removal rate and efficiency. The aminated straw-assisted PAO1 to increase the phenol degraded concentration from 1900 to 1,500 mg/L, and shorten the degraded time by 44 h at 1,500 mg/L phenol. The immobilized PAO1 could remove phenol at pH 10 and 11, which was 2.7 and 3.8 times higher than free bacteria, respectively. In addition, the immobilized PAO1 could totally remove phenol, which was twice as high as that of free bacteria, at 4% NaCl stress. Furthermore, the removal efficiency of immobilized PAO1 was higher than that of free bacteria under the stress of various metal ions, especially for Co2+ and Pb2+. The determination coefficients R2 and root mean square error showed that the back propagation artificial neural network model could predict the degradation of phenol under various conditions, saving time and economic cost. The present study can be envisioned that this biomaterial has great potential in the bioremediation of organic pollutions.

Comprehensive maps of material stock dynamics reveal increasingly coordinated urban development in the Yangtze River Delta of China

Sustainable urban development critically depends on effectively managing the interplay between material stock (MS) and economic growth. This study combined convolutional neural network model and nighttime lights data to map building MS of Yangtze River Delta (YRD) urban agglomeration in China from 2000 to 2020 across 1 km × 1 km pixel scale, then uncovered the spatiotemporal dynamics of MS and its correlation with economic development. Our findings indicate that the model performed robustly on the test set (R2 > 0.88). YRD's MS surged over tenfold, reaching 20,772 teragram, primarily expanding along northwest-southeast developmental axes. Most YRD cities exhibited synchronized growth in material stock and GDP, suggesting an emergent pattern of sustainable urban expansion. However, cities at the developmental extremes highlighted the need for optimizing urban development strategies. By categorizing YRD cities into four distinct development modes, our study offers deep insights into the dynamics of urban development, underpinning targeted strategies that could guide cities towards more sustainable and resource-efficient growth trajectories.

Design and Optimization of Key Parameters for a Machine Vision-Based Walnut Shell–Kernel Separation Device

The separation of walnut kernels and shells has long been regarded as a bottleneck, limiting processing efficiency, product quality, and industry advancement. In response to the challenges of improving separation accuracy and the inadequacy of existing equipment for meeting industry demands, this paper proposes an innovative walnut shell–kernel separation device based on machine vision technology. An experimental system was constructed, and key parameters were optimized to enhance its performance. The device comprises five main modules: material conveyance, image acquisition, control module, sorting module, and frame. Differential separation technology is used to convert the walnut material group into a stable particle flow, addressing the issue of missed selections due to material blockages. An enhanced YOLOv8n algorithm improves small object detection and interference resistance, enabling accurate identification of walnut kernels. The Box–Behnken Design and Artificial Neural Network prediction model was used to determine the optimal operating parameters for the device. Experimental results showed that effective differential separation was achieved when the dual-stage conveyor system operated at speeds of 0.2 m/s and 1 m/s. The improved I-YOLOv8n algorithm reached an accuracy of 98.8%. Using the neural network model, the optimal operational parameters were determined: an air pressure of 0.72 MPa, a jetting component angle of 10.16°, and a sorting height of 105.12 cm. Under these conditions, the device achieved an actual cleaning rate of 93.56%, demonstrating outstanding separation performance. Compared to traditional separation methods, this device offers significant advantages in terms of efficiency, quality, and ease of operation, providing new technological pathways and support for the automation and intelligent transformation of the walnut processing industry. In the future, the device is expected to undergo further improvements to meet broader market demand and serve as a reference for the separation of other agricultural products.

Cell‐preserving scheme for mechanobiological research on dedifferentiation of chondrocytes

AbstractIn the present work, we introduce a scheme for efficient mechanobiological research on the dedifferentiation of chondrocytes (CHs) using a deep neural network (DNN) model that does not require sacrificing the cells and thus, saving resources. Cells are a part of active biological systems subjected to physical stimuli such as mechanical loading. Such loading affects cellular processes including proliferation, differentiation, and the interplay with the surrounding environment. CHs are mechanosensitive cells that produce and maintain the cartilaginous matrix that allows cartilage to bear and distribute mechanical loads in joints; their role in cartilage regeneration and treatment of osteoarthritis (OA) has been the focus of research projects. Nevertheless, it remains an open challenge to fully understand the effect of mechanical stimuli on CHs. One of the unresolved mechanobiological behaviors of CHs is dedifferentiation. Dedifferentiation of CHs is the phenomenon where isolated CHs alter their phenotype when cultured in a 2D in vitro environment over passages. While intact, cobblestone‐shaped CHs mainly produce collagen II, dedifferentiated CHs with elongated shapes produce fibroblastic collagen I. This limits the scalability of a promising treatment for OA, called ‘autologous chondrocyte implantation (ACI)’. To overcome this limit, research is being carried out to understand the dedifferentiation mechanism in detail. This proposed scheme is composed of three parts: (i) the cell‐seeded specimens are loaded in a bioreactor system, (ii) an optical microscope is used to obtain the phase‐contrast images of living cells, and (iii) the images are analyzed using a DNN. This scheme provides an efficient tool to analyze the ratio of intact to dedifferentiated CHs while preserving cells on our way to elucidate the effects of mechanical loading on the dedifferentiation of CHs.

Inelastic scattering of PO+ by H2 at interstellar temperatures

Abstract Phosphorous species are of great interest in interstellar chemistry since they are the basic blocks for building life here on Earth. Modelling the abundance and environment of recently detected PO+ under non Local Thermodynamic Equilibrium (LTE) requires rotational spectra of the molecule along with accurate collisional rates with the most abundant species, hydrogen and helium. A new 4D ab initio potential energy surface (PES) of PO+ - H2 collision is calculated using CCSD(T)/CBS(DTQ) methodology considering rigid rotor approximation. The region containing the minima of the PES is augmented using neural networks (NN) model while very high potentials (>2500 cm−1) and asymptotic region have been approximated using Slater and R−4 functions respectively. The close coupling calculations have been performed using MOLSCAT software for both ortho and para-H2. The rate coefficients have been reported for transitions j − j′ = 1 − 0, 2 − 1, 3 − 2 and 5 − 4 through which PO+ has been experimentally detected in ISM. The rate coefficients for even and odd transitions of PO+ with para-H2 are compared with that of helium and are found to be 1.1-2.0 times higher. For even transitions (Δj = 2), the ortho-H2 rates are 10% higher than para-H2 rates. However, the trend reverses in the case of odd transitions (Δj = 1) when higher J transitions are considered at low temperatures. At higher temperatures, the ortho rates cross the para-H2 rates and become larger than the latter. The new rate coefficients with both ortho and para-H2 will enable accurate modelling of the PO+ abundance in the interstellar medium under non-LTE conditions.

Efficient Solutions to Time-Fractional Telegraph Equations with Chebyshev Neural Networks

Abstract The objective of this study is to employ artificial neural networks (ANNs) as a new method for solving time fractional telegraph equations (TFTEs), which are typically tackled using the Caputo fractional derivative in scientific investigations. The utilisation of Chebyshev polynomials in place of a hidden layer serves to improve computing performance and expand the range of input patterns. A feed-forward neural network (NN) model is utilised, using Adam optimisation, to further refine network parameters and minimise mistakes. Two problems are selected, and the proposed method is employed to obtain their numerical solutions in order to showcase the effectiveness of the methodology. The efficiency of the proposed method is demonstrated by a comparison with the analytical solutions.

Breast Tumor Diagnosis Based on Molecular Learning Vector Quantization Neural Networks.

DNA nanotechnology plays a crucial role in precise cancer medicine. Currently, molecular logic circuits are applied to detect tumor-specific biomarkers and control the release of therapeutic drugs. However, these systems lack self-learning capabilities for intelligent diagnostics in biological samples, and their data processing capabilities are limited. Here, a molecular learning vector quantization neural network (LVQNN) model based on DNA strand displacement (DSD) technology for breast tumor diagnosis is developed. Compared to previous work, the molecular LVQNN boasts powerful computing abilities, handling high-dimensional data for intelligent cancer diagnosis. To verify the feasibility and versatility of the network, two distinct typical datasets are selected: one from a single source with cell morphology data from 569 cases, and a more extensive one spanning different populations and ages, with miRNA gene expression data from 1881 cases. By using the molecular LVQNN, diagnostic experiments are conducted on 50 and 120 public individuals from these two datasets, respectively, achieving accuracy rates of 94% and 97.5%. This study demonstrates that the LVQNN model exhibits significant advantages in breast cancer diagnosis and enhances diagnostic accuracy while introducing new approaches for intelligent cancer diagnosis, anticipated to bring significant breakthroughs and application prospects to precise cancermedicine.

Motor imagery EEG signal classification based on deformable convolution v3 and adaptive spatial attention mechanism

This paper is dedicated to solving the issue of insufficient feature extraction and declining model performance in the classification of motor imagery electroencephalogram (EEG) signals. We propose an enhanced ResNet18 neural network model and introduce two pivotal innovations at the model level. Firstly, we present the deformable convolution v3 (DCNv3), an improved convolutional structure that overcomes the limitations of the original deformable convolution. Through unbiased design with center sampling points, DCNv3 achieves more precise feature localization, enhances the model’s representation capability for complex structures, and reduces the risk of overfitting. Simultaneously, it comprehensively captures features in effective information regions, thereby strengthening the model’s generalization ability. Furthermore, this enhancement design also improves computational efficiency, reducing computational and parameter burdens. Secondly, we introduce a novel Adaptive Spatial Attention Mechanism (ASAM) aimed at precisely capturing feature correlations, thereby avoiding information loss. We conducted extensive comparative experiments on publicly available datasets BCI IV dataset 2a, BCI IV dataset 2b, and BCI III dataset 3a, achieving accuracy rates of 91.32%, 92.3%, and 90.47% respectively. Compared to other methods, our approach not only significantly improves the average classification accuracy but also demonstrates remarkable generalization capabilities. These results validate the effectiveness and stability of our method, thereby better meeting the practical application requirements of brain-computer interface technology.

Research on Failure Pressure Prediction of Water Supply Pipe Based on GA-BP Neural Network

The water supply pipeline is regarded as the “lifeline” of the city. In recent years, pipeline accidents caused by aging and other factors are common and have caused large economic losses. Therefore, in order to avoid large economic losses, it is necessary to analyze the failure prediction of pipelines so that the pipelines that are going to fail can be replaced in a timely manner. In this paper, we propose a method for predicting the failure pressure of pipelines, i.e., a genetic algorithm was used to optimize the weights and thresholds of a BP neural network. The first step was to determine the topology of the neural network and the number of input and output variables. The second step was to optimize the weights and thresholds initially set for the back propagation neural network using a genetic algorithm. Finally, the optimized back-propagation neural network was used to simulate and predict pipeline failures. It was proved by examples that compared with the separate back propagation neural network model and the optimized and trained genetic algorithm-back propagation neural network, the model performed better in simulation prediction, and the prediction accuracy could reach up to 91%, whereas the unoptimized back propagation neural network model could only reach 85%. It is feasible to apply this model for fault prediction of pipelines.

AutoPeptideML: A study on how to build more trustworthy peptide bioactivity predictors.

Automated machine learning (AutoML) solutions can bridge the gap between new computational advances and their real-world applications by enabling experimental scientists to build their own custom models. We examine different steps in the development life-cycle of peptide bioactivity binary predictors and identify key steps where automation can not only result in a more accessible method, but also more robust and interpretable evaluation leading to more trustworthy models. We present a new automated method for drawing negative peptides that achieves better balance between specificity and generalisation than current alternatives. We study the effect of homology-based partitioning for generating the training and testing data subsets and demonstrate that model performance is overestimated when no such homology correction is used, which indicates that prior studies may have overestimated their performance when applied to new peptide sequences. We also conduct a systematic analysis of different protein language models as peptide representation methods and find that they can serve as better descriptors than a naive alternative, but that there is no significant difference across models with different sizes or algorithms. Finally, we demonstrate that an ensemble of optimised traditional machine learning algorithms can compete with more complex neural network models, while being more computationally efficient. We integrate these findings into AutoPeptideML, an easy-to-use AutoML tool to allow researchers without a computational background to build new predictive models for peptide bioactivity in a matter of minutes. Source code, documentation, and data are available at https://github.com/IBM/AutoPeptideML and a dedicated web-server at http://peptide.ucd.ie/AutoPeptideML. Raul.fernandezdiaz@ucdconnect.ie or denis.shields@ucd.ie. Supplementary Information can be accessed from Zenodo: Https://zenodo.org/records/13363975.

Unconscious classification of quantitative electroencephalogram features from propofol versus propofol combined with etomidate anesthesia using one-dimensional convolutional neural network

ObjectiveEstablishing a convolutional neural network model for the recognition of characteristic raw electroencephalogram (EEG) signals is crucial for monitoring consciousness levels and guiding anesthetic drug administration.MethodsThis trial was conducted from December 2023 to March 2024. A total of 40 surgery patients were randomly divided into either a propofol group (1% propofol injection, 10 mL: 100 mg) (P group) or a propofol-etomidate combination group (1% propofol injection, 10 mL: 100 mg, and 0.2% etomidate injection, 10 mL: 20 mg, mixed at a 2:1 volume ratio) (EP group). In the P group, target-controlled infusion (TCI) was employed for sedation induction, with an initial effect site concentration set at 5–6 μg/mL. The EP group received an intravenous push with a dosage of 0.2 mL/kg. Six consciousness-related EEG features were extracted from both groups and analyzed using four prediction models: support vector machine (SVM), Gaussian Naive Bayes (GNB), artificial neural network (ANN), and one-dimensional convolutional neural network (1D CNN). The performance of the models was evaluated based on accuracy, precision, recall, and F1-score.ResultsThe power spectral density (94%) and alpha/beta ratio (72%) demonstrated higher accuracy as indicators for assessing consciousness. The classification accuracy of the 1D CNN model for anesthesia-induced unconsciousness (97%) surpassed that of the SVM (83%), GNB (81%), and ANN (83%) models, with a significance level of p < 0.05. Furthermore, the mean and mean difference ± standard error of the primary power values for the EP and P groups during the induced period were as follows: delta (23.85 and 16.79, 7.055 ± 0.817, p < 0.001), theta (10.74 and 8.743, 1.995 ± 0.7045, p < 0.02), and total power (24.31 and 19.72, 4.588 ± 0.7107, p < 0.001).ConclusionLarge slow-wave oscillations, power spectral density, and the alpha/beta ratio are effective indicators of changes in consciousness during intravenous anesthesia with a propofol-etomidate combination. These indicators can aid anesthesiologists in evaluating the depth of anesthesia and adjusting dosages accordingly. The 1D CNN model, which incorporates consciousness-related EEG features, represents a promising tool for assessing the depth of anesthesia.Clinical Trial Registrationhttps://www.chictr.org.cn/index.html.

Hybrid modeling for in silico optimization of a dynamic perfusion cell culture process.

The bio-pharmaceutical industry heavily relies on mammalian cells for the production of bio-therapeutic proteins. The complexity of implementing and high cost-of-goods of these processes are currently limiting more widespread patient access. This is driving efforts to enhance cell culture productivity and cost reduction. Upstream process intensification (PI), using perfusion approaches in the seed train and/or the main bioreactor, has shown substantial promise to enhance productivity. However, developing optimal process conditions for perfusion-based processes remain challenging due to resource and time constraints. Model-based optimization offers a solution by systematically screening process parameters like temperature, pH, and culture media to find the optimum conditions in silico. To our knowledge, this is the first experimentally validated model to explain the perfusion dynamics under different operating conditions and scales for process optimization. The hybrid model accurately describes Chinese hamster ovary (CHO) cell culture growth dynamics and a neural network model explains the production of mAb, allowing for optimization of media exchange rates. Results from six perfusion runs in Ambr® 250 demonstrated high accuracy, confirming the model's utility. Further, the implementation of dynamic media exchange rate schedule determined through model-based optimization resulted in 50% increase in volumetric productivity. Additionally, two 5 L-scale experiments validated the model's reliable extrapolation capabilities to large bioreactors. This approach could reduce the number of wet lab experiments needed for culture process optimization, offering a promising avenue for improving productivity, cost-of-goods in bio-pharmaceutical manufacturing, in turn improving patient access to pivotal medicine.

Recognition of vehicle license plates in highway scenes with deep fusion network and connectionist temporal classification

AbstractLicense plate recognition is crucial in Intelligent Transportation Systems (ITS) for vehicle management, traffic monitoring, and security inspection. In highway scenarios, this task faces challenges such as diversity, blurriness, occlusion, and illumination variation of license plates. This article explores Recurrent Neural Networks based on Connectionist Temporal Classification (RNN‐CTC) for license plate recognition in challenging highway conditions. Four neural network models: ResNet50, ResNeXt, InceptionV3, and SENet, all combined with RNN‐CTC are comparatively evaluated. Furthermore, a novel architecture named ResNet50 Deep Fusion Network using Connectionist Temporal Classification (ResNet50‐DFN‐CTC) is proposed. Comparative and ablation experiments are conducted using the Highway License Plate Dataset of Southeast University (HLPD‐SU). Results demonstrate the superior performance of ResNet50‐DFN‐CTC in challenging highway conditions, achieving 93.158% accuracy with a processing time of 7.91 ms, outperforming other tested models. This research contributes to advancing license plate recognition technology for real‐world highway applications under adverse conditions.

Machine learning insights into the evolution of flood Resilience: A synthesized framework study

Enhancing urban resilience represented a viable strategy to mitigate flooding induced by intense human activities and climate change. However, existing studies often concentrated on system attributes or isolated resilience characteristics, failing to offer a holistic evaluation of urban flood resilience performance. Thus, it was imperative to develop a comprehensive flood resilience framework that incorporated the resilience evolution process including resistance, economic and function recovery. Consequently, this study endeavored to devise a synthesized framework for evaluating urban flood resilience, subsequently employing a Convolutional Neural Network (CNN) model for simulation. The findings indicated that: (1) Guangzhou’s maximum resistance capacity diminished from 0.52 to 0.50 as rainfall return periods altered, while Dongguan exhibited the lowest resistance, decreasing from 0.42 to 0.40. Regarding functional recovery capacity, Guangzhou ranked highest (0.35) and Foshan lowest (0.19); (2) according to Triangular Fuzzy Number-based AHP (TFN-AHP) analysis, the area classified as highest in resilience decreased from 15.6% to 12.1% of the total, whereas the low resilience area increased from 7.6% to 8.7%; (3) Zhuhai and Zhaoqing were primarily clustered along the resistance axis, in contrast, Dongguan was distinguished by its advancement along the axis of functional recovery.(4) CNN simulations yielded precise outcomes, with the Area Under the Receiver Operating Characteristic Curve (AUC) and predictive accuracy (ACC) values exceeding 0.8,respectively. The insights provided by this research were crucial for entities tasked with flood risk management.

Construction of an artificial neural network diagnostic model and investigation of immune cell infiltration characteristics for idiopathic pulmonary fibrosis

BackgroundIdiopathic pulmonary fibrosis (IPF) is a severe lung condition, and finding better ways to diagnose and treat the disease is crucial for improving patient outcomes. Our study sought to develop an artificial neural network (ANN) model for IPF and determine the immune cell types that differed between the IPF and control groups.MethodsFrom the Gene Expression Omnibus (GEO) database, we first obtained IPF microarray datasets. To conduct protein-protein interaction (PPI) networks and enrichment analyses, differentially expressed genes (DEGs) were screened between tissues of patients with IPF and tissues of controls. Afterward, we identified the important feature genes associated with IPF using random forest (RF) analysis, and then constructed and validated a prediction ANN mode. In addition, the proportions of immune cells were quantified using cell-type identification by estimating relative subsets of RNA transcripts (CIBERSORT) analysis, which was performed on microarray datasets based on gene expression profiling.ResultsA total of 11 downregulated and 36 upregulated DEGs were identified. PPI networks and enrichment analyses were carried out; the immune system and extracellular matrix were the subjects of the enrichments. Using RF analysis, the significant feature genes LRRC17, COMP, ASPN, CRTAC1, POSTN, COL3A1, PEBP4, IL13RA2, and CA4 were identified. The nine feature gene scores were integrated into the ANN to develop a diagnostic prediction model. The receiver operating characteristic (ROC) curves demonstrated the strong diagnostic ability of the ANN in predicting IPF in the training and testing sets. An analysis of IPF tissues in comparison to normal tissues revealed a reduction in the infiltration of natural killer cells resting, monocytes, macrophages M0, and neutrophils; conversely, the infiltration of T cells CD4 memory resting, mast cells, and macrophages M0 increased.ConclusionLRRC17, COMP, ASPN, CRTAC1, POSTN, COL3A1, PEBP4, IL13RA2, and CA4 were determined as key feature genes for IPF. The nine feature genes in the ANN model will be extremely important for diagnosing IPF. It may be possible to use differentiated immune cells from IPF samples in comparison to normal samples as targets for immunotherapy in patients with IPF.

Seismic response prediction for buildings with sliding live loads using neural networks

The seismic response of primary structures (PS) can be significantly influenced by live loads, particularly when these loads consist of stacked bodies that are capable of sliding during seismic events. Conventional design methods often fail to account for the energy dissipation resulting from such sliding, leading to overly conservative structural estimates. This study examines the effects of sliding live loads on the seismic response of a PS, where two rigid bodies, referred to as secondary bodies (SBs), are stacked as live loads on a PS. The interaction between the lower SB and the PS, as well as between the SBs, is modelled using Coulomb's friction model. The governing equations are solved using the fourth-order Runge-Kutta method. Seismic Zones III and V from IS 1893:2016 are considered as hazard levels. A parametric investigation explores the impact of varying dynamic parameters of both the PS and SBs on the seismic response. The results demonstrate significant energy dissipation within the stack due to sliding, which, if not considered, results in conservative displacement estimates. The study also evaluates how friction coefficients, mass ratios, and excitation levels affect the stack's energy dissipation capacity. A novel method to compute the modified primary structural period (Tnew) for design purposes is proposed, and an artificial neural network (ANN) model is developed, achieving a high prediction accuracy (R2 = 99 %). Sensitivity analysis is performed to assess the influence of input parameters on the output. This research highlights the need to include sliding loads in seismic design.

Development of message passing-based graph convolutional networks for classifying cancer pathology reports

BackgroundApplying graph convolutional networks (GCN) to the classification of free-form natural language texts leveraged by graph-of-words features (TextGCN) was studied and confirmed to be an effective means of describing complex natural language texts. However, the text classification models based on the TextGCN possess weaknesses in terms of memory consumption and model dissemination and distribution. In this paper, we present a fast message passing network (FastMPN), implementing a GCN with message passing architecture that provides versatility and flexibility by allowing trainable node embedding and edge weights, helping the GCN model find the better solution. We applied the FastMPN model to the task of clinical information extraction from cancer pathology reports, extracting the following six properties: main site, subsite, laterality, histology, behavior, and grade.ResultsWe evaluated the clinical task performance of the FastMPN models in terms of micro- and macro-averaged F1 scores. A comparison was performed with the multi-task convolutional neural network (MT-CNN) model. Results show that the FastMPN model is equivalent to or better than the MT-CNN.ConclusionsOur implementation revealed that our FastMPN model, which is based on the PyTorch platform, can train a large corpus (667,290 training samples) with 202,373 unique words in less than 3 minutes per epoch using one NVIDIA V100 hardware accelerator. Our experiments demonstrated that using this implementation, the clinical task performance scores of information extraction related to tumors from cancer pathology reports were highly competitive.