Joint Network Research Articles

Exercising and immuno-oncology treatment (ExIO): a prospective pilot study for patients with solid tumors.

Despite exercise's known benefits for cancer patients, little is understood about combining it with immune checkpoint inhibitors (ICI). This study aims to assess the feasibility and acceptability of a personalized physical activity program for cancer patients on ICI treatment and to investigate its effects on quality of life, ICI-related side effects, fatigue, and emotional symptoms. This prospective study was conducted across eight cancer practices in a joint network in Brazil. Eligible cancer patients undergoing ICIs were recruited from June 2022 to January 2023. Participants completed a 12-week physical activity program with remote supervision. Baseline and 12-week assessments utilized the FACT-ICM, BFI, ESAS, PROMIS-Anxiety, and FCR-7 scales. Feasibility and acceptability were evaluated based on enrollment, retention, and adherence rates, with patient feedback collected at week 12. Linear mixed models were employed to assess outcomes. Out of 38 patients, 81% showed high adherence, 62% engaged in activities more than 4days weekly, and 97% were satisfied with the program. The sample was mostly female (55%), median age of 60.5, primarily diagnosed with breast cancer (26%) and renal cell carcinoma (24%), with 60.5% having metastatic cancer. Sixty percent received ICI combined with chemotherapy. Significant improvements were observed over time in treatment-related side effects, emotional symptoms, and quality of life (p < 0.001). These findings endorse integrating an exercise program with remote supervision into the care of cancer patients undergoing ICI treatment. This intervention may positively impact symptom management and quality of life, offering a cost-effective approach in oncologic treatment and survivorship care globally.

Trajectories of human brain functional connectome maturation across the birth transition.

Understanding the sequence and timing of brain functional network development at the beginning of human life is critically important from both normative and clinical perspectives. Yet, we presently lack rigorous examination of the longitudinal emergence of human brain functional networks over the birth transition. Leveraging a large, longitudinal perinatal functional magnetic resonance imaging (fMRI) data set, this study models developmental trajectories of brain functional networks spanning 25 to 55 weeks of post-conceptual gestational age (GA). The final sample includes 126 fetal scans (GA = 31.36 ± 3.83 weeks) and 58 infant scans (GA = 48.17 ± 3.73 weeks) from 140 unique subjects. In this study, we document the developmental changes of resting-state functional connectivity (RSFC) over the birth transition, evident at both network and graph levels. We observe that growth patterns are regionally specific, with some areas showing minimal RSFC changes, while others exhibit a dramatic increase at birth. Examples with birth-triggered dramatic change include RSFC within the subcortical network, within the superior frontal network, within the occipital-cerebellum joint network, as well as the cross-hemisphere RSFC between the bilateral sensorimotor networks and between the bilateral temporal network. Our graph analysis further emphasized the subcortical network as the only region of the brain exhibiting a significant increase in local efficiency around birth, while a concomitant gradual increase was found in global efficiency in sensorimotor and parietal-frontal regions throughout the fetal to neonatal period. This work unveils fundamental aspects of early brain development and lays the foundation for future work on the influence of environmental factors on this process.

Two-stage dual-attention spatiotemporal joint network model for multi-energy load prediction of integrated energy system

Accurate prediction of multi-energy load is essential for the scheduling, safe operation, and stability of integrated energy systems. This paper proposes a novel two-stage multi-energy load prediction model with dual-attention spatiotemporal joint network for integrated energy system. In the first stage of multi-task prediction, the integrated self-attention long short-term memory (LSTM) network is built to forecast the multi-energy load initially, and an error correction model on basis of BP network is constructed to reduce the impact of error accumulation on the second stage prediction. In the second stage of single-task prediction, a parallel spatiotemporal joint network model is proposed to extract the static spatial feature and dynamic temporal feature between multi-energy loads and between loads and meteorological factors, which fully captures the complex interactions and dependencies between the multi-energy loads. The prediction model integrates self-attention and graph attention mechanisms in both stages to improve the feature extraction capabilities. The simulation results demonstrate that, compared to bidirectional LSTM, LSTM-gated recurrent unit, time convolutional networks-LSTM and variational mode decomposition-convolutional neural network-LSTM models, the proposed multi-energy load prediction model reduces the average absolute percentage error by 66.25%, 16.83%, 53.58% and 5.29%, respectively.

Extension of a 3D Geological Entity Modeling Method for Discontinuous Deformation Analysis

AbstractWith the development of 3D discontinuous deformation analysis (DDA) in precise stress fields and crack propagation problems, it has also demonstrated outstanding capabilities in solving continuous–discontinuous problems. However, currently, 3D DDA modeling primarily focuses on generating rock joint networks and developing 3D cutting algorithms. Correspondingly, 3D geological modeling methods are not yet mature, and establishing 3D models often demands substantial time. The lack of supporting preprocessing modeling methods and corresponding visual operation interfaces significantly hampers the development of 3D DDA. This method builds upon advanced research achievements in unmanned aerial vehicle oblique photography, 3D reconstruction, 3D cutting, computer graphics, and visualization program design. This research establishes a 3D geological entity modeling method for 3D DDA and constructs a comprehensive program using relevant C++ libraries and C language interfaces. In this method, a 3D geological model that incorporates geological elements such as strata and faults is initially established using non‐uniform rational B‐splines (NURBSs) surfaces as the boundary of the solid model. Subsequently, finite element meshing is applied, followed by corresponding topology transformation, resulting in a 3D block system model suitable for 3D DDA calculation. To cater for diverse application scenarios, continuous–discontinuous models integrated with subblocks and models of arbitrary polyhedra can be established. The proposed method has been validated through several typical modeling examples, showing its ability to rapidly and generate 3D high‐precision geological reality models suitable for 3D DDA calculations. Additionally, some techniques used in this method can be extended for modeling other numerical simulation methods, warranting further research.

Joint Network Expansion Planning With Internet Data Center and Charging Stations for EV Sharing System

Joint Network Expansion Planning With Internet Data Center and Charging Stations for EV Sharing System

An Attention-Based Spatial-Spectral Joint Network for Maize Hyperspectral Images Disease Detection

Maize is susceptible to pest disease, and the production of maize would suffer a significant decline without precise early detection. Hyperspectral imaging is well-suited for the precise detection of diseases due to its ability to capture the internal chemical characteristics of vegetation. However, the abundance of redundant information in hyperspectral data poses challenges in extracting significant features. To overcome the above problems, in this study we proposed an attention-based spatial-spectral joint network model for hyperspectral detection of pest-infected maize. The model contains 3D and 2D convolutional layers that extract features from both spatial and spectral domains to improve the identification capability of hyperspectral images. Moreover, the model is embedded with an attention mechanism that improves feature representation by focusing on important spatial and spectral-wise information and enhances the feature extraction ability of the model. Experimental results demonstrate the effectiveness of the proposed model across different field scenarios, achieving overall accuracies (OAs) of 99.24% and 97.4% on close-up hyperspectral images and middle-shot hyperspectral images, respectively. Even under the condition of a lack of training data, the proposed model performs a superior performance relative to other models and achieves OAs of 98.29% and 92.18%. These results proved the validity of the proposed model, and it is accomplished efficiently for pest-infected maize detection. The proposed model is believed to have the potential to be applied to mobile devices such as field robots in order to monitor and detect infected maize automatically.

Super-resolution enhancement and segmentation for digital rock based on multi-task joint deep neural network

Reliable segmentation of pores and minerals from high-resolution (HR) digital rock data is the fundamental prerequisite for accurately characterizing the digital rock's physical properties. Limited by the complexity of the super-resolution (SR) issues and the user bias for segmentation, conventional SR enhancement and segmentation methods are generally brutal in satisfying the research demands. In recent years, with the introduction of deep learning technology into digital rock research, deep learning-based approaches for SR enhancement or segmentation methods have emerged and achieved significantly better effects than conventional approaches. Most deep learning-based digital rock SR enhancement and segmentation processing steps are currently separate. Still, preliminary studies have demonstrated that an end-to-end approach that integrates the two steps could achieve better performance. However, the training cost of 3D deep neural networks that can be directly applied to 3D digital rock processing is usually expensive. In contrast, 2D networks cannot be effectively applied to SR segmentation of 3D digital rocks. Here, we present an end-to-end SR segmentation framework for 3D digital rock based on a 2D multi-task joint deep neural network. The multi-task joint networks utilize a parallel architecture that integrates the SegNet for the segmentation task and the EDSR for the SR task. We also improved the loss function to address the issue of category imbalance and proposed an approach to utilize the 2D network for 3D digital rock processing. We demonstrate the effectiveness of the proposed network framework, the improved loss function, and the 3D digital rock processing strategies through ablation experiments on the high and low-resolution (LR) CT image datasets captured by imaging devices. The results show that the evaluation metrics and physical properties of the SR-segmented 2D\\3D digital rocks align with the HR-segmented results (ground truth). It indicates that the proposed framework can improve the performance of digital rock SR enhancement and segmentation and that it is essential to integrate deep learning frameworks into digital rock analysis.

Shape-Guided Detection: A joint network combining object detection and underwater image enhancement together

Most of the existing underwater image object detection methods involve pre-processing, such as using underwater image enhancement, to improve the accuracy of object detection. However, pre-processing methods are designed to improve the subjective perception of the human eye, which does not necessarily improve the object detection performance and consumes a large amount of computational resources. Therefore, in this paper, we creatively combine these two tasks and propose a Shape-Guided Detection network (SGD) to simultaneously optimize underwater image enhancement and object detection. In the SGD network, we innovatively incorporate the prior shape features as a learnable module embedded in it to fully explore the shape characteristics and structural details of the target object. To ensure that the prior knowledge can be effectively fused into the global network structure, we design a Shape Prior Enhancement module, which aims to realize the deep integration of the prior information with the local details. In order to optimize the stability of model training and enhance its convergence performance, a dual strategy of explicit and implicit constraints is ingeniously proposed in our method. We conduct extensive experiments on public datasets and the results show that the combination of our method with different detectors significantly improves the performance. The object detection performance reaches up to 0.491 mAP for optical images and 0.576 mAP for sonar images, and improves the preprocessing speed by 0.1 s.

A Joint Network for Low-Light Image Enhancement Based on Retinex

Methods based on the physical Retinex model are effective in enhancing low-light images, adeptly handling the challenges posed by low signal-to-noise ratios and high noise in images captured under weak lighting conditions. However, traditional models based on manually designed Retinex priors do not adapt well to complex and varying degradation environments. DEANet (Jiang et al., Tsinghua Sci Technol. 2023;28(4):743–53 2023) combines frequency and Retinex to address the interference of high-frequency noise in low-light image restoration. Nonetheless, low-frequency noise still significantly impacts the restoration of low-light images. To overcome this issue, this paper integrates the physical Retinex model with deep learning to propose a joint network model, DEANet++, for enhancing low-light images. The model is divided into three modules: decomposition, enhancement, and adjustment. The decomposition module employs a data-driven approach based on Retinex theory to split the image; the enhancement module restores degradation and adjusts brightness in the decomposed images; and the adjustment module restores details and adjusts complex features in the enhanced images. Trained on the publicly available LOL dataset, DEANet++ not only surpasses the control group in both visual and quantitative aspects but also achieves superior results compared to other Retinex-based enhancement methods. Ablation studies and additional experiments highlight the importance of each component in this method.

FINANCIAL AND ECONOMIC PROSPECTS FOR CREATING GREEN ENERGY PROJECTS IN THE UKRAINE-POLAND COLLABORATION

This paper examines the economic and financial advantages of improving international collaboration between Ukraine and Poland in the development of alternative energy projects. It conducts a thorough evaluation of existing and forthcoming renewable energy generation technologies, using expert, Delphi, and methods based on statistics to identify the most promising and universally applicable solutions. The calculation of the payback period for small-scale green energy projects pays particular attention to the actual duration of the period in the case of small business ventures. The findings are summarized with a vision for the future Ukrainian-Polish cooperation in the scope of green energy cooperation and specific instruments and processes that could contribute to the success of the collaboration.The article outlines the potential approach that involves the isolation of the cooperation between Ukraine and Poland in green energy relying on the creation of investment and venture funds, public support for the manufacturing of a renewable energy component, and the establishment of non-volatile industrial complexes. Additionally, measures such as independent small alternative power plants and the initiatives for the development of the Ukrainian-Polish joint network of alternative energy are mentioned.The current war between Ukraine and Russia has drawn attention to the crucial need for energy security and diversification, making Ukraine-Poland collaboration on green energy projects all the more important. Ukraine may minimize its reliance on the Russian energy supply by focusing on renewable energy programs, while Poland can invest in sustainable energy to boost regional stability and economic progress. The article points out that the continuation and advancement of green energy initiatives require international cooperation. The paper offers detailed guidance on how the financial, economic, and technological opportunities and relations between Ukraine and Poland may be enhanced.

Directional fracture patterns of excavated jointed rock mass within rough discrete fractures

The influence of the fractures on the anisotropic mechanical behavior of excavated fractured rock mass is important for the stability of rock tunnels. Rough joints were generated using fractal principles, and the Monte Carlo method was employed to create joint network faces. Biaxial compression numerical simulations were conducted using the discrete element method to analyze the behavior of the rock mass containing these networks. Quantitative analysis was carried out to evaluate the strength variability and total number of cracks in the jointed rock mass. A linear regression analysis was conducted to establish the relationship between rock strength and excavation size. The findings demonstrate significant disparities between the linear and fractal models regarding rock damage mode, stress–strain curves, and crack evolution patterns. Tensile rupture and local shear damage were found to be the primary fracture modes for the fractured rock tunnels, forming a V-shaped damage zone within the central tunnel. The shape of this zone was influenced by fracture direction and excavation ratio. Excavation at the center of the rock mass substantially impacted the variability of rock strength and the occurrence of internal cracks. An increase in the slope ratio led to a greater variability of rock mass strength, transitioning from nearly isotropic to weakly anisotropic behavior. The total number of cracks exhibited a W-shaped trend, initially increasing and later decreasing as the slope ratio decreased. Additionally, a clear linear correlation was observed between rock strength and excavation ratio at the center of the rock mass.

Predicting the Stability of Rock Slopes in the Presence of Diverse Joint Networks and External Factors Using Machine Learning Algorithms

Predicting the Stability of Rock Slopes in the Presence of Diverse Joint Networks and External Factors Using Machine Learning Algorithms

“Our similarities are different” The relationship between alexithymia and depression

Alexithymia is a multi-faceted personality trait, which is the inability to recognize and describe emotions. It is associated with a multitude of mental health problems, and its implication for the diagnosis and treatment of depression remains unclear. The current study explored the nuances of the relationship between alexithymia and depression in a sample of 210 patients with depression. We assessed alexithymia with the 20-Item Toronto Alexithymia Scale (TAS-20) and depression with the Beck Depression Inventory (BDI-I). The mean TAS-20 score was 57.47 ± 10.63, and the mean BDI-I score was 49.33±9.24. We explored the network structure of alexithymia and depression. Items related to difficulties in identifying, describing, and expressing feelings were prominent in the alexithymia network. Joy, guilt, and self-dislike stand out in the depression network. In our analysis, we were able to show the crescent relationship between depression and alexithymia, with an inflection point at a TAS-20 score of 53. Although the correlation-concordance index was moderate (0.41; 95 %CI: 0.29–0.51), both scales greatly overlap. In the joint network of alexithymia and depression, we could identify bridge (i.e., connecting) items between alexithymia and depression. These were difficulties understanding and relating feelings to physical and body sensations on the alexithymia side, and self-dislike, crying, and somatic concern on the depression side. Taken together, they point to the pivotal role of alexithymia in the somatization/embodiment of emotions and feelings in depression.

Naturally fractured reservoir characterisation in heterogeneous sandstones: insight for uranium in situ recovery (Imouraren, Niger)

Abstract. This study delves into the characterisation of a heterogenous reservoir, the Tchirezrine II sandstone unit in northern Niger. The characterisation is crucial for potential uranium in situ recovery (ISR) in a naturally fractured and faulted context. Employing a multifaceted approach, including well log data, optical borehole imagery, and hydrogeological tests, alongside satellite-based lineament analysis, this study provides a comprehensive understanding of the structures and their impact on fluid flow. Lineament analysis reveals scale-dependent patterns, consistent with spatially homogeneous joint networks restricted to mechanical units, as well as nearly scale-invariant patterns, better corresponding to spatially heterogeneous fault networks. Various deformation structures are detected from borehole imagery, including extensional fractures, cataclastic deformation bands, and brecciated–cataclastic fault cores. Based on well log data, the Tchirezrine II reservoir displays heterogeneous porosity and permeability related to its fluvial context. These data differ from the traditional porosity–permeability relationship obtained in a sandstone reservoir matrix but are instead consistent with Nelson's classification, emphasising the impact of deformation structures on such petrophysical properties. Hydrological tests have been implemented into a zone of E–W-trending deformation structures, revealing a strong permeability anisotropy. This strong E–W anisotropy is consistent with the presence of the observed E–W structures, i.e. with a draining behaviour of extensional open fractures and a sealing behaviour of both cataclastic bands and fault rocks. Considering implications for ISR mining, this study allows the discussion of the interplay between fractures, faults, and fluid flow properties. It suggests that a well pattern perpendicular to the main permeability orientation can attenuate channelled flow, thus improving the contact of the leach solution with the mineralised matrix. These results provide an integrated approach and a multi-scale characterisation of naturally fractured reservoir (NFR) properties in sandstone, offering a basis for the optimisation of NFR production such as ISR development.

A Joint Network of Edge-Aware and Spectral-Spatial Feature Learning for Hyperspectral Image Classification.

Hyperspectral image (HSI) classification is a vital part of the HSI application field. Since HSIs contain rich spectral information, it is a major challenge to effectively extract deep representation features. In existing methods, although edge data augmentation is used to strengthen the edge representation, a large amount of high-frequency noise is also introduced at the edges. In addition, the importance of different spectra for classification decisions has not been emphasized. Responding to the above challenges, we propose an edge-aware and spectral-spatial feature learning network (ESSN). ESSN contains an edge feature augment block and a spectral-spatial feature extraction block. Firstly, in the edge feature augment block, the edges of the image are sensed, and the edge features of different spectral bands are adaptively strengthened. Then, in the spectral-spatial feature extraction block, the weights of different spectra are adaptively adjusted, and more comprehensive depth representation features are extracted on this basis. Extensive experiments on three publicly available hyperspectral datasets have been conducted, and the experimental results indicate that the proposed method has higher accuracy and immunity to interference compared to state-of-the-art (SOTA) method.

An integrated approach for characterization of a fractured-rock carbonate aquifer in the Zagros Region of Iran

Karst aquifers typically exhibit a wide range of dissolution effects that include end-members of matrix, fractured-rock, and conduit-dominated types. This study employs an integrated approach involving geological studies, hydrogeological assessments, hydrochemical and isotopic analysis, and dye tracer tests to investigate the Sarvak limestone aquifer (SLA) in the Zagros Region, southwest Iran. Key characteristics of the SLA include numerous stratification boundaries, thin limestone layers with intercalated siliceous and marl impurities, extensive joint and fracture networks, predominant autogenic recharge, and absence of notable point recharge features (e.g., sinkholes, dolines), and the exchange flow with the Bakhtiari River (B-R) has made it a unique aquifer. Numerous pieces of evidence, such as low spatial and temporal changes in groundwater level, insignificant seasonal variations in the hydrochemical and isotopic composition of water samples, and the supersaturation state of groundwater with calcite and dolomite minerals, suggested a slow flow regime in many parts of the SLA. This regime is characterized by low velocity and long residence time of groundwater. The results reveal that despite the high solubility of carbonate rocks, extensive joint networks can limit significant karst development, leading to fluid flow behavior similar to that of fractured-rock aquifers. Therefore, SLA can be considered a fractured-rock and karst aquifer. The identified paths with fast flow in the dam area are unremarkable and cannot provide a large portion of the Pelle Khan Spring discharge.

A deep learning-based approach for predicting coking duration in a commercial coke oven

ABSTRACT The prediction of coking duration is critical during operation of commercial coke oven batteries; however, it is still far from accurately predict since it has to rely on experience-based estimation. To address this problem, we proposed a machine learning-based approach, and presented a general practice and theoretical framework for predicting coking duration accurately and efficiently. Three types of neutral networks, including the multi-layer perceptron (MLP) model, the radial basis function (RBF) model, and the long short-term memory (LSTM) model, were tested and evaluated in terms of predicting accuracy, training and predicting time. On this basis, a novel two-stream joint MLP-RBF neural network was proposed. All the networks were trained and tested with onsite-collected datasets. In the modeling, the potential relationships between variables are deeply learnt automatically. The results show that the innovative joint network inherits advantages from both sides and exhibits the lowest prediction error, 2.79% (MAPE) or 4.26% (RMSPE), and is recommended as the most appropriate for practical application. This work not only presents an efficient approach for predicting coking duration, but also examines the extent to which the machine learning technology can be employed during the digital transformation of traditional industries.

UFNet: Joint U-Net and Fully Connected Neural Network to Bias Correct Precipitation Predictions from Climate Models

Abstract Precipitation values produced by climate models are biased due to the parameterization of physical processes and limited spatial resolution. Current bias-correction approaches usually focus on correcting lower-order statistics (mean and standard deviation), which make it difficult to capture precipitation extremes. However, accurate modeling of extremes is critical for policymaking to mitigate and adapt to the effects of climate change. We develop a deep learning framework, leveraging information from key dynamical variables impacting precipitation to also match higher-order statistics (skewness and kurtosis) for the entire precipitation distribution, including extremes. The deep learning framework consists of a two-part architecture: a U-Net convolutional network to capture the spatiotemporal distribution of precipitation and a fully connected network to capture the distribution of higher-order statistics. The joint network, termed UFNet, can simultaneously improve the spatial structure of the modeled precipitation and capture the distribution of extreme precipitation values. Using climate model simulation data and observations that are climatologically similar but not strictly paired, the UFNet identifies and corrects the climate model biases, significantly improving the estimation of daily precipitation as measured by a broad range of spatiotemporal statistics. In particular, UFNet significantly improves the underestimation of extreme precipitation values seen with current bias-correction methods. Our approach constitutes a generalized framework for correcting other climate model variables which improves the accuracy of the climate model predictions, while utilizing a simpler and more stable training process.

Two guidance joint network based on coarse map and edge map for camouflaged object detection

Two guidance joint network based on coarse map and edge map for camouflaged object detection

Simultaneous clustering and estimation of networks in multiple graphical models.

Gaussian graphical models are widely used to study the dependence structure among variables. When samples are obtained from multiple conditions or populations, joint analysis of multiple graphical models are desired due to their capacity to borrow strength across populations. Nonetheless, existing methods often overlook the varying levels of similarity between populations, leading to unsatisfactory results. Moreover, in many applications, learning the population-level clustering structure itself is of particular interest. In this article, we develop a novel method, called Simultaneous Clustering and Estimation of Networks via Tensor decomposition (SCENT), that simultaneously clusters and estimates graphical models from multiple populations. Precision matrices from different populations are uniquely organized as a three-way tensor array, and a low-rank sparse model is proposed for joint population clustering and network estimation. We develop a penalized likelihood method and an augmented Lagrangian algorithm for model fitting. We also establish the clustering accuracy and norm consistency of the estimated precision matrices. We demonstrate the efficacy of the proposed method with comprehensive simulation studies. The application to the Genotype-Tissue Expression multi-tissue gene expression data provides important insights into tissue clustering and gene coexpression patterns in multiple brain tissues.