Correlation Of Features Research Articles

Cracking and Packing Information about the Features of Expected Rewards in the Orbitofrontal Cortex.

The orbitofrontal cortex (OFC) is crucial for tracking various aspects of expected outcomes, thereby helping to guide choices and support learning. Our previous study showed that the effects of reward timing and size on the activity of single units in OFC were dissociable when these attributes were manipulated independently ( Roesch et al., 2006). However, in real-life decision-making scenarios, outcome features often change simultaneously, so here we investigated how OFC neurons in male rats integrate information about the timing and identity (flavor) of reward and respond to changes in these features, according to whether they were changed simultaneously or separately. We found that a substantial number of OFC neurons fired differentially to immediate versus delayed reward and to the different reward flavors. However, contrary to the previous study, selectivity for timing was strongly correlated with selectivity for identity. Taken together with the previous research, these results suggest that when reward features are correlated, OFC tends to "pack" them into unitary constructs, whereas when they are independent, OFC tends to "crack" them into separate constructs. Furthermore, we found that when both reward timing and flavor were changed, reward-responsive OFC neurons showed unique activity patterns preceding and during the omission of an expected reward. Interestingly, this OFC activity is similar and slightly preceded the ventral tegmental area dopamine (VTA DA) activity observed in a previous study ( Takahashi et al., 2023), consistent with the role of OFC in providing predictive information to VTA DA neurons.

Generative adversarial network for Multimodal Contrastive Domain Sharing based on efficient invariant feature-centric growth analysis improved brain tumor classification

ABSTRACT Efficient and accurate classification of brain tumor categories remains a critical challenge in medical imaging. While existing techniques have made strides, their reliance on generic features often leads to suboptimal results. To overcome these issues, Multimodal Contrastive Domain Sharing Generative Adversarial Network for Improved Brain Tumor Classification Based on Efficient Invariant Feature Centric Growth Analysis (MCDS-GNN-IBTC-CGA) is proposed in this manuscript.Here, the input imagesare amassed from brain tumor dataset. Then the input images are preprocesssed using Range – Doppler Matched Filter (RDMF) for improving the quality of the image. Then Ternary Pattern and Discrete Wavelet Transforms (TPDWT) is employed for feature extraction and focusing on white, gray mass, edge correlation, and depth features. The proposed method leverages Multimodal Contrastive Domain Sharing Generative Adversarial Network (MCDS-GNN) to categorize brain tumor images into Glioma, Meningioma, and Pituitary tumors. Finally, Coati Optimization Algorithm (COA) optimizes MCDS-GNN’s weight parameters. The proposed MCDS-GNN-IBTC-CGA is empirically evaluated utilizing accuracy, specificity, sensitivity, Precision, F1-score,Mean Square Error (MSE). Here, MCDS-GNN-IBTC-CGA attains 12.75%, 11.39%, 13.35%, 11.42% and 12.98% greater accuracy comparing to the existingstate-of-the-arts techniques, likeMRI brain tumor categorization utilizing parallel deep convolutional neural networks (PDCNN-BTC), attention-guided convolutional neural network for the categorization of braintumor (AGCNN-BTC), intelligent driven deep residual learning method for the categorization of braintumor (DCRN-BTC),fully convolutional neural networks method for the classification of braintumor (FCNN-BTC), Convolutional Neural Network and Multi-Layer Perceptron based brain tumor classification (CNN-MLP-BTC) respectively.

Revolutionizing engineered cementitious composite materials (ECC): the impact of XGBoost-SHAP analysis on polyvinyl alcohol (PVA) based ECC predictions

This study integrates previous experimental data and employs machine learning (ML) methods, including Random Forest (RF), Support Vector Machine (SVM), Artificial Neural Network (ANN), and eXtreme Gradient Boosting (XGBoost), to predict the compressive strength (CS) and tensile strength (TS) of engineered cementitious composites (ECC). XGBoost emerged as the superior model among the four ML models, providing an interpretable and highly accurate predictive framework. To optimize the model performance, hyperparameter tuning using a fivefold cross-validation approach with the data divided into 80% training and 20% testing subsets. The Shapley Additive Explanations (SHAP) algorithm was also employed to reveal the impact of important features, such as the water/binder ratio, fly ash content, and water reducer dosage, on the model’s predictions and their interrelationships. The XGBoost demonstrates the most exemplary performance, as reflected in the R2 values of 0.92 and 0.97 for CS and TS testing, respectively. The SHAP analysis provided insights into the impact of individual features on CS and TS, shedding light on how specific characteristics influence the predictive accuracy of these properties. This highly accurate prediction model uncovers insights into correlated features, aids in creating new mix designs of ECC, and supports global efforts toward a low-carbon future in the construction industry by reducing carbon emissions.

Enhancing Intercultural Business English Communication Factors Evaluation System Using the Termite Life Cycle Optimization Algorithm and Dynamically Stabilized Recurrent Neural Network

In today's globalized business environment, effective intercultural communication in English is paramount for successful collaboration among professionals from diverse backgrounds. To enhance the accuracy of the evaluation system, enhancing intercultural business English communication factors evaluation system using the termite life cycle optimization algorithm and dynamically stabilized recurrent neural network (IBEC–DSRNN–TLCOA) is proposed in this manuscript. The input image is captured from mobile camera. Then the input images are preprocessed using intelligent weight decreasing firefly–particle filtering (IWDFPF) to remove noise and enhance the input images. Afterwards, the preprocessed image is fed to the entropy-founded spatial fuzzy C-means clustering approach for segmenting the image. Then the contrast, correlation, energy and homogeneousness features are extracted by using force-invariant improved feature extraction technique. The extracted features are given to dynamically stabilized recurrent neural network (DSRNN) to image target detection and English description generation. Termite life cycle optimization algorithm (TLCOA) is employed to enhance the weight parameters of DSRNN. The proposed IBEC–DSRNN–TLCOA method is implemented. The proposed IBEC–DSRNN–TLCOA method provides 32.53%, 31.86%, and 35.72% higher accuracy; 35.58%, 32.16%, and 37.72% higher F-measure when compared with the existing methods, such as exploration of intelligent translation with evaluation systems for business English (IBEC–RCNN), E-learning engagement with convolution neural networks on business education (IBEC–CNN), and deep neural network-based research on scoring business English oral training (IBEC–DNN), respectively.

A weighted time embedding transformer network for remaining useful life prediction of rolling bearing

Data-driven remaining useful life (RUL) prediction is of vital importance to industrial equipment prognostics health management (PHM). The transformer algorithms have been applied for RUL prediction recently. RUL prediction is a typically time-correlated task. However, in most transformer-based RUL prediction methods, one of the main procedures, patch embedding, is usually implemented by linear-based or convolution-based approaches, which lacks consideration for time correlation feature extraction. Moreover, the time correlation features at different degradation states may be variable, which will make different contributions to RUL prediction. To this end, in this paper, a novel framework for RUL prediction named weighted time embedding transformer (WTE-Trans) is proposed. A WTE module is designed to enhance the capacity of the model to extract more discriminative time correlation features and impose different weights automatically. Concretely, the peak-to-peak values are calculated from input and transformed firstly to provide prior knowledge information, and then multiplied by a learnable mask to generate an adaptive weight. Afterwards, the adaptive weight is used to impose constraints on the transformed input. Secondly, multiple 3D convolution (3D-conv) modules are designed to integrate the deep features through the time axis. Finally, the shifted-window transformer block is adopted to complete the RUL prediction task. Experiments are carried out on the widely used PHM 2012 bearing dataset and an actual industrial bearing full-lifecycle dataset to verify the effectiveness and application value. The results and analysis have shown that lower prediction error can be achieved by the proposed method.

Precipitation nowcasting leveraging spatial correlation feature extraction and deep spatio-temporal fusion network

Precipitation nowcasting leveraging spatial correlation feature extraction and deep spatio-temporal fusion network

Rheo-Optical Near-Infrared (NIR) Characterization of Styrene-Butadiene Rubber (SBR) Using Two-Trace Two-Dimensional (2T2D) Correlation Analysis.

A rheo-optical characterization technique based on the combination of near-infrared (NIR) spectroscopy and tensile testing was applied for the first time to an actual rubber sample based on styrene-butadiene rubber (SBR) including silica filler. When SBR samples were subjected to mechanical deformation, changes in the NIR spectral features were readily captured. Two-trace two-dimensional (2T2D) correlation analysis was then applied to the sets of NIR spectra to clearly reveal the subtle but pertinent difference between the NIR spectral features of the initial and deformed SBR. The initial deformation of the sample induces greater deformation of the soft butadiene groups than of the hard styrene groups. The inclusion of the silica filler and a coupling agent (CA) essentially develops firm links between the silica and butadiene via the CA to restrict the displacement of the butadiene during the tensile deformation of the system. The development of such linkage requires even more mechanical force to deform the SBR, which, in turn, improves Young's modulus of the rubber system. Asynchronous correlation spectra of SBR with no silica filler revealed that, during the deformation of the SBR, the butadiene groups were initially deformed, and this feature was then replaced by the predominant deformation of the hard styrene groups. On the other hand, this correlation feature became somewhat unclear when a similar analysis was applied to the SBR sample with silica filler, revealing subtle differences in interaction between individual comonomer functional groups distributed randomly along the copolymer chain and CA.

OPTIMIZED U-NET SEGMENTATION MODEL AND DEEP MAXOUT CLASSIFIER FOR BRAIN TUMOR CLASSIFICATION

The most serious nervous system ailment, a brain tumor impairs one’s health seriously and ultimately results in death. MRI, one of the most frequently used medical imaging modalities for brain tumors, has emerged as the main diagnostic system for the treatment and study of brain tumors. It was challenging to segment and classify the many kinds of brain tumors. The swarm intelligence approach has the potential to more efficiently and effectively tackle a number of issues. Therefore, this work develops a novel model for the classification of brain tumors that includes various phases. Primarily, the input image is preprocessed via the proposed median filtering that aids in removing noises. Subsequently, segmentation is done via optimal U-Net. For precise segmentation, the weights are tuned optimally by battle royale optimization with the Bernoulli randomization (BROBR) algorithm. Then, features like the proposed local Gabor XOR pattern (PLGXP), texton features, gray level co-occurrence matrix (GLCM), and correlation features are extracted. Finally, BTC is done using a deep maxout network (DMO) that provides the final output on the absence or presence of a brain tumor.

Fault transformer: An automatic fault detection algorithm on seismic images using a transformer enhanced neural network

Seismic fault detection is a key step in seismic interpretation and reservoir characterization that often requires a large amount of human labor and interpretation time. Therefore, automatic seismic fault detection is critical for improving the efficiency of seismic data processing and interpretation. Existing artificial intelligence methods are mostly based on convolutional neural networks with a U-shaped encoder-decoder structure, known as U-net. However, the convolution is limited in modeling long-range correlative features. Instead, transformers, using self-attention mechanisms, avoid the local nature of the convolution, which has the potential to extract long-distance correlations. Transformers are proven to perform well in natural language processing, image classification, and segmentation tasks in precision and recall. Here, we develop a new deep neural network with transformers and a U-net-like structure: a fault transformer to perform the fault detection task. The new network outperforms the traditional U-net in the application with synthetic data sets.

FORMICARY SWARM OPTIMIZED DEEP CNN FOR FACIAL EMOTION RECOGNITION FROM HUMAN FACIAL EXPRESSIONS

Facial emotion recognition (FER) is a dominant research area that captures the biological facial features and matches the facial data with existing databases to analyze the individual’s emotional state. Numerous techniques have been formulated for attaining effective FER. However, the occlusions, different head positions, deformed faces, motion blur under unrestricted settings, and complicated backgrounds make it complex to analyze the facial images. In this paper, the formicary swarm optimization-based deep convolutional neural network (FSO-opt DCNN) model is utilized for facial emotion detection for which JAFFE and the RAVDESS facial expression datasets are used. DCNNs are proficient with built-in feature extraction strategies from images to map the various facial expressions to the corresponding emotional states adopted for effective FER. In addition, the intensity, directional, and edge patterns as well as the correlation features extracted utilizing the hybrid textual pattern, RESNET 101 and VGG 16-based correlation modules assist the DCNN to attain the informative feature from the high-resolution images. Further, the formicary swarm optimization (FSO) is incorporated that effectively tunes the DCNN to capture the complex relationships between the learned features that excel the FER capability. Evaluating the metrics, the face recognition model using the RAVDESS dataset achieves notable efficiencies during a training percentage (TP) of 90%, with values of 97.51%, 95.48%, 99.55%, 95.48%, 97.48%, 96.47%, and a minimum loss of 2.49%. Simultaneously, the emotion detection model demonstrates robust efficiencies with metric values of 96.75%, 98.49%, 95.01%, 98.49%, 96.72%, 97.59%, and a minimum loss of 3.25%. Finally, the obtained results reveal the efficacy of the FSO-opt DCNN, particularly in face recognition tasks, as it outperforms the existing models across various datasets, showcasing its versatility and potential in facial analysis applications.

WiVelo: Fine-grained Wi-Fi Walking Velocity Estimation

Passive human tracking using Wi-Fi has been researched broadly in the past decade. Besides straightforward anchor point localization, velocity is another vital sign adopted by the existing approaches to infer user trajectory. However, state-of-the-art Wi-Fi velocity estimation relies on Doppler-Frequency-Shift (DFS), which suffers from the inevitable signal noise incurring unbounded velocity errors, further degrading the tracking accuracy. In this article, we present WiVelo, which explores new spatial-temporal signal correlation features observed from different antennas to achieve accurate velocity estimation. First, we use subcarrier shift distribution (SSD) extracted from channel state information (CSI) to define two correlation features for direction and speed estimation, separately. Then, we design a mesh model calculated by the antennas’ locations to enable a fine-grained velocity estimation with bounded direction error. Finally, with the continuously estimated velocity, we develop an end-to-end trajectory recovery algorithm to mitigate velocity outliers with the property of walking velocity continuity. We implement WiVelo on commodity Wi-Fi hardware and extensively evaluate its tracking accuracy in various environments. The experimental results show our median and 90-percentile tracking errors are 0.47 m and 1.06 m, which are half and a quarter of state-of-the-art. The datasets and source codes are published through Github ( https://github.com/research-source/code ).

Fault diagnosis of induction motor in the cooling water supply system using a multi-channel data fusion transformer with limited sample conditions*

Abstract Induction motors play a vital role in the cooling water supply system of hydropower facilities. However, it is not feasible to collect sufficient fault samples in a hydropower station. The scarcity of labeled samples poses a challenge in developing powerful diagnostic models with high classification accuracy. To address this challenge, this paper proposes a multi-channel data fusion strategy based on a transformer for feature enhancement. Initially, the original signals are transferred into non-overlapping single-channel data patches to preserve correlation features across different channels. Next, temporal and spatial attention modules are applied to process the data patches, which can learn and fuse temporal and spatial information, respectively. Subsequently, the data patches are embedded to retain position information and represent fault-related features through class embedding, which are further processed by a transformer encoder with self-attention mechanisms. Finally, the classification task is achieved by using a multilayer perceptron layer connected to the class embedding. While dealing with limited training samples, the proposed method can learn robust features that are beneficial to improve the fault diagnosis ability of induction motors. The comparison of the proposed method with three basic models and two improved methods demonstrates the superiority of the proposed method in accuracy and feature clustering performance under limited sample conditions. In addition, ablation experiments demonstrate the necessity of each module in the proposed method.

Method for determining the types of encryption of linear block codes

Problem statement. All known methods for determining the parameters of the encrypted code are based on known types of channel encoding and encryption, but the scrambling code data received by the uncoordinated communication parties is unknown, therefore it is necessary to analyze whether the outgoing signal is encoded, and which encryption method is adopted, contributing to further parameter determination. Goal. Development of a methodology for determining the types of encryption of linear block codes by combining correlation features and convolutional neural networks. Results. The simulation results show that compared with the traditional feature detection method, this method, based on a convolutional neural network and long short-term memory, has a better recognition speed at BER and can recognize the type of encryption of linear block code in BER and noise conditions. Practical significance. The identification of the encryption code parameters is used to subsequently determine the parameters of the encoder and has significant practical engineering value. The work was carried out with the financial support of the Ministry of Science and Higher Education of the Russian Federation within the framework of the state assignment "youth laboratory" No. FZGM-2024-0003.

Diagnosing Breast Cancer Using AI: A Comparison of Deep Learning and Traditional Machine Learning Methods

Breast cancer remains a significant health concern globally, with early detection being crucial for effective treatment. In this study, we explore the predictive power of various diagnostic features in breast cancer using machine learning techniques. We analyzed a dataset comprising clinical measurements of mammograms from 569 patients, including mean radius, texture, perimeter, area, and smoothness, alongside the diagnosis outcome. Our methodology involves preprocessing steps such as handling missing values and removing duplicates, followed by a correlation analysis to identify and eliminate highly correlated features. Subsequently, we train eight machine learning models, including Logistic Regression (LR), K-Nearest Neighbors (K-NN), Linear Support Vector Machine (SVM), Kernel SVM, Naïve Bayes, Decision Trees Classifier (DTC), Random Forest Classifier (RFC), and Artificial Neural Networks (ANN), to predict the diagnosis based on the selected features. Through comprehensive evaluation metrics such as accuracy and confusion matrices, we assess the performance of each model. Our findings reveal promising results, with 6 out of 8 models achieving high accuracy (>90%), with ANN having the highest accuracy in diagnosing breast cancer based on the selected features. These results underscore the potential of machine learning algorithms in aiding early breast cancer diagnosis and highlight the importance of feature selection in improving predictive performance.

Novel feature selection method for accurate breast cancer classification using Correlation coefficient and Modified GWO Algorithm

Breast cancer is perceived as the most common cause of mortality among women globally. Early detection of this disease is critical to reduce significantly the possibility of death. Machine learning techniques have been proved to be efficient and very successful for an accurate breast cancer diagnosis. In this paper, an efficient hybrid Feature Selection (FS) method named a Correlation technique-Modified Grey Wolf Optimizer (CMGWO) was proposed for accurate breast cancer classification based on dimensionality reduction. The suggested technique is based on two stages: the feature selection step and the classification step. Feature selection is the process of picking the most significant characteristics from a dataset. This stage is crucial in machine learning. Firstly, we focus on the filter method by using a Correlation technique for dimensionality reduction. This technique is intended to eliminate and reduce the number of features by selecting one feature from the other correlated features. Secondly, we use the Modified Grey Wolf Optimization algorithm (MGWO) to locate and determine the most significant features from uncorrelated features. After that, we use multiple classifiers to classify breast cancer disease based on the selected features. The Wisconsin Diagnostic Breast Cancer (WDBC) database was used to prove the performance of our proposed work. The experimental results show that the combination of the correlation method and MGWO for feature selection increases the accuracy rate of classification with a minimum number of features. The performances of different machine learning algorithms were evaluated, including Random Forest classifier (RF), Support Vector Machine (SVM) Classifier, and Naïve Bayes (NB) Classifier for the classification step. The suggested technique proves to be the best approach and reliable one among all studied approaches since it increases classification accuracy to 99.12\% obtained by CMGWO using Random Forest classifier and demonstrates its significance in detecting breast cancer.

Magnetic Resonance Imaging Features of Sporadic Optic Chiasmatic-Hypothalamic Gliomas and Correlation with Histopathology and BRAF Gene Alterations.

Sporadic optic chiasmatic-hypothalamic gliomas (OCHGs), though histologically low-grade tumors, manifest as aggressive neoplasms radiologically, leading to difficulty in diagnosis. Molecular alterations of the BRAF gene are detectable in a majority of sporadic OCHGs. The purpose of our study was to elucidate the characteristic imaging features of sporadic OCHGs and to investigate whether imaging phenotypes could potentially correlate with specific BRAF gene alterations associated with these tumors. We retrospectively reviewed baseline magnetic resonance (MR) images and medical records of 26 patients with histopathologically proven sporadic OCHGs. MR imaging (MRI) features were systematically evaluated. Statistical analysis was performed to determine whether there was a significant association between imaging findings and BRAF molecular alterations. Twenty-two cases (84.6%) presented with solid-cystic masses, while four (15.4%) presented with purely solid lesions. In all 26 cases, the solid component revealed central necrosis; there was minimal necrosis in 11 cases (42.3%), moderate in 8 (30.7%), and marked in 7 (26.9%). The presence of multiple cysts (>4) and minimal necrosis showed a significant association with BRAFV600E mutation (P < 0.005). Marked necrosis in the solid component significantly correlated with BRAF wild genotype (P < 0.001). The presence of a single peripheral cyst significantly correlated with BRAF fusion (P = 0.04). Sporadic OCHGs have a distinctive appearance on imaging. The solid-cystic composition coupled with varying degrees of central necrosis are clues to the radiological diagnosis of this entity and can facilitate early recognition in clinical practice. Imaging could potentially serve as a non-invasive predictor of the BRAF alteration status, thereby serving as a prognostic marker and guiding personalized management.

Characteristic features of strong correlation: lessons from a 3-fermion one-dimensional harmonic trap

Abstract Many quantum phenomena responsible for key applications in material science and quantum chemistry arise in the strongly correlated regime. This is at the same time, a costly regime for computer simulations. In the limit of strong correlation analytic solutions exist, but as we move away from this limit numerical simulation are needed, and accurate quantum solutions do not scale well with the number of interacting particles. In this work we propose to use few-particle harmonic traps in combination with twisted light as a quantum emulator to investigate the transition into a strongly-correlated regime. Using both analytic derivations and numerical simulations we generalize previous findings on 2 Coulomb interacting fermions trapped in a one-dimensional harmonic trap to the case of 3 fermions. The 4 signatures of strong correlation we have identified in the one-dimensional harmonic trap are: (i) the ground state density is highly localized around N equilibrium positions, where N is the number of particles, (ii) the symmetric and antisymmetric ground state wavefunctions become degenerate, (iii) the von Neumann entropy grows, (iv) the energy spectrum is fully characterized by N normal modes or less. Our findings describe the low-energy behavior of electrons in quantum wires and ions in Paul traps. Similar features have also been reported for cold atoms in optical lattices.

A knowledge-data integration framework for rolling element bearing RUL prediction across its life cycle

Prediction of Remaining Useful Life (RUL) for Rolling Element Bearings (REB) has attracted widespread attention from academia and industry. However, there are still several bottlenecks, including the effective utilization of multi-sensor data, the interpretability of prediction models, and the prediction across the entire life cycle, which limit prediction accuracy. In view of that, we propose a knowledge-based explainable life-cycle RUL prediction framework. First, considering the feature fusion of fast-changing signals, the Pearson correlation coefficient matrix and feature transformation objective function are incorporated to an Improved Graph Convolutional Autoencoder. Furthermore, to integrate the multi-source signals, a Cascaded Multi-head Self-attention Autoencoder with Characteristic Guidance is proposed to construct health indicators. Then, the whole life cycle of REB is divided into different stages based on the Continuous Gradient Recognition with Outlier Detection. With the development of Measurement-based Correction Life Formula and Bidirectional Recursive Gated Dual Attention Unit, accurate life-cycle RUL prediction is achieved. Data from self-designed test rig and PHM 2012 Prognostic challenge datasets are analyzed with the proposed framework and five existing prediction models. Compared with the strongest prediction model among the five, the proposed framework demonstrates significant improvements. For the data from self-designed test rig, there is a 1.66 % enhancement in Corrected Cumulative Relative Accuracy (CCRA) and a 49.00 % improvement in Coefficient of Determination (R2). For the PHM 2012 datasets, there is a 4.04 % increase in CCRA and a 120.72 % boost in R2.

Development of a High-Throughput Pipeline to Characterize Microglia Morphological States at a Single-Cell Resolution.

As rapid responders to their environments, microglia engage in functions that are mirrored by their cellular morphology. Microglia are classically thought to exhibit a ramified morphology under homeostatic conditions which switches to an ameboid form during inflammatory conditions. However, microglia display a wide spectrum of morphologies outside of this dichotomy, including rod-like, ramified, ameboid, and hypertrophic states, which have been observed across brain regions, neurodevelopmental timepoints, and various pathological contexts. We applied dimensionality reduction and clustering to consider contributions of multiple morphology measures together to define a spectrum of microglial morphological states in a mouse dataset that we used to demonstrate the utility of our toolset. Using ImageJ, we first developed a semiautomated approach to characterize 27 morphology features from hundreds to thousands of individual microglial cells in a brain region-specific manner. Within this pool of features, we defined distinct sets of highly correlated features that describe different aspects of morphology, including branch length, branching complexity, territory span, and circularity. When considered together, these sets of features drove different morphological clusters. Our tools captured morphological states similarly and robustly when applied to independent datasets and using different immunofluorescent markers for microglia. We have compiled our morphology analysis pipeline into an accessible, easy-to-use, and fully open-source ImageJ macro and R package that the neuroscience community can expand upon and directly apply to their own analyses. Outcomes from this work will supply the field with new tools to systematically evaluate the heterogeneity of microglia morphological states across various experimental models and research questions.

Ultra-short-term wind power forecasting method based on multi-variable joint extraction of spatial-temporal features

Accurate and reliable wind power forecasting is imperative for wind power stations' stable and efficient operation. Information such as wind speed and wind direction in the same wind field has spatial-temporal differences. Considering the spatial-temporal changes in wind fields can improve model prediction accuracy. However, existing methods suffer from limited ability to capture correlation features among variables, information loss in spatial-temporal feature extraction, and neglect short-term temporal features. This paper introduces a novel ultra-short-term wind power forecasting method based on the combination of a deep separable convolutional neural network (DSCNN) and long- and short-term time-series network (LSTNet), incorporating maximum information coefficient (MIC) to realize multi-variable joint extraction of spatial-temporal features. The method utilizes MIC to jointly analyze and process the multi-variate variables before spatial-temporal feature extraction to avoid information redundancy. The spatial features between input variables and wind power are extracted by deep convolution and pointwise convolution in DSCNN. Then, a convolutional neural network and gated recurrent unit in LSTNet are combined to capture long-term and short-term temporal features. In addition, an autoregressive module is employed to accept features extracted by MIC to enhance the model's learning of temporal features. Based on real datasets, the performance of models is validated through comprehensive evaluation experiments such as comparison experiments, ablation experiments, and interval prediction methods. The results show that the proposed method reduces mean absolute error by up to 4.66% and provides more accurate prediction intervals, verifying the accuracy and effectiveness of the proposed method.