Aggregate Characteristics Research Articles

Evaluating dimensionality reduction of comorbidities for predictive modeling in individuals with neurofibromatosis type 1.

Dimensionality reduction techniques aim to enhance the performance of machine learning (ML) models by reducing noise and mitigating overfitting. We sought to compare the effect of different dimensionality reduction methods for comorbidity features extracted from electronic health records (EHRs) on the performance of ML models for predicting the development of various sub-phenotypes in children with Neurofibromatosis type 1 (NF1). EHR-derived data from pediatric subjects with a confirmed clinical diagnosis of NF1 were used to create 10 unique comorbidities code-derived feature sets by incorporating dimensionality reduction techniques using raw International Classification of Diseases codes, Clinical Classifications Software Refined, and Phecode mapping schemes. We compared the performance of logistic regression, XGBoost, and random forest models utilizing each feature set. XGBoost-based predictive models were most successful at predicting NF1 sub-phenotypes. Overall, features based on domain knowledge-informed mapping schema performed better than unsupervised feature reduction methods. High-level features exhibited the worst performance across models and outcomes, suggesting excessive information loss with over-aggregation of features. Model performance is significantly impacted by dimensionality reduction techniques and varies by specific ML algorithm and outcome being predicted. Automated methods using existing knowledge and ontology databases can effectively aggregate features extracted from EHRs. Dimensionality reduction through feature aggregation can enhance the performance of ML models, particularly in high-dimensional datasets with small sample sizes, commonly found in EHRs health applications. However, if not carefully optimized, it can lead to information loss and data oversimplification, potentially adversely affecting model performance.

An HVS-derived network for assessing the quality of camouflaged targets with feature fusion

High-value assets on the battlefield typically require adequate camouflage to evade detection and annihilation by enemy scouts. Consequently, artificial camouflage technology is extensively acknowledged and utilized as a crucial defensive tactic in the military sphere. The quality of camouflage performance was assessed by military observers through the human visual system (HVS). This method involved locating the camouflaged objects and rating the camouflaged degree against the background. Current camouflage assessment methods typically involved the manual extraction and aggregation of objective features throughout an image. These approaches fall short in constructing a correlation mapping between objective features and subjective perceptions of camouflaged objects, culminating in imprecise assessments and discrepancies. To address these issues, this paper presents the first three-stage full-reference learning framework for locating camouflaged objects, extracting camouflage features, and assessing camouflage quality. Given the lack of datasets specifically designed for evaluating camouflage quality, we have contributed a datasets focused on human-camouflaged targets. The experimental results show that the three-stage framework is remarkably accurate in assessing the camouflage quality, leading to an explainable network. The camouflaged people quality assessment(CPQA) dataset is available at http://github.com/samsunq/CPQA_Datasets.git.

Optimizing thermal and strength characteristics of loess-based ceramic aggregates for sustainable buildings

Optimizing thermal and strength characteristics of loess-based ceramic aggregates for sustainable buildings

Drug toxicity prediction model based on enhanced graph neural network.

Prediction of drug toxicity remains a significant challenge and an essential process in drug discovery. Traditional machine learning algorithms struggle to capture the full scope of molecular structure features, limiting their effectiveness in toxicity prediction. Graph Neural Network offers a promising solution by effectively extracting drug features from their molecular graphs. However, existing graph learning algorithms fail to account for the interaction features between graph nodes and the indirect edges connecting them. This paper proposes an enhanced graph Neural Network algorithm that employs multi-view features for each node, capturing the feature interactions between each node and its neighbors. Additionally, the adjacency matrix is preprocessed to handle indirect edge interactions. A pooling technique is then applied to aggregate node features, followed by normalization and an activation layer. To further enhance the proposed algorithm, multi-scale attention is applied to learn graph features at different scales, utilizing weights to understand intricate relationships among node feature vectors. The proposed algorithm is evaluated using eight toxicity datasets, covering binary classification, multi-task multi-class, and regression tasks. For binary classification, the Tox21, AMES, Skin reaction, Carcinogens, and DILI datasets are tested. For multi-task multi-class, the ToxCast dataset is applied, and for regression, the LD50 and hREG datasets are tested. The proposed algorithm is compared with four well-known algorithms including Graph Convolution Network, Graph Attention Network, Graph Isomorphism Network, Enhanced Graph Isomorphism Network, and Graph Total Variation. For the classification task, the proposed algorithm achieves ROC-AUC scores of 0.752 for Tox21, 0.775 for AMES, 0.707 for Skin reaction, 0.845 for Carcinogens, 0.92 for DILI, and 0.691 for the ToxCast dataset. For the regression task, the algorithm attains mean square errors of 0.896 for the LD50 dataset and 0.766 for the hREG dataset. These results demonstrate an improvement over the compared algorithms across all evaluated datasets.

EPFDNet: Camouflage object detection with edge perception in frequency domain

Camouflaged object detection (COD) is a relatively new field of computer vision research. The challenge of this task lies in accurately segmenting camouflaged objects from backgrounds that are similar in appearance. In fact, the generation of reliable edges is an effective mean of distinguishing between the foreground and background of the image, which is beneficial for assisting in determining the location of camouflaged objects. Inspired by this, we design an Edge Encoder that decomposes features into different frequency bands adopting learnable wavelets and focuses on high-frequency components with sufficient edge details to extract accurate edges. Subsequently, the Feature Aggregation Module is proposed to integrate contextual features, which generates rough edge details by sensing the difference between two branch features and use this information to further refine our edge features. Furthermore, the Stage Enhancement Module is developed to enhance the features through reverse attention guidance and dilate convolution, which mines the detailed structural information of the camouflaged objects area by eliminating foreground. The superiority of our proposed method (EPFDNet) over the existing 17 state-of-the-art methods is demonstrated through extensive experiments on three widely used COD benchmark datasets. The code has been released at https://github.com/LitterMa-820/EPFDNet.

Automatic crack defect detection via multiscale feature aggregation and adaptive fusion

Automatic crack defect detection via multiscale feature aggregation and adaptive fusion

Mechanism underlying the effect of soluble oat β-glucan and tea polyphenols on wheat gluten aggregation characteristics.

The mechanism of how the coexistence of oat β-glucan (OβG) and tea polyphenols (TP) impacts gluten aggregation properties was investigated. The OβG might form interchain hydrogen bondings and compete for water with gluten, which could increase gluten aggregation and the gluten network's expansion, leading to its increasing average particle size (by 17.23 %) with 5%OβG. The physicochemical characteristics of TP and OβG + TP groups showed similar changing trends, indicating the predominant effect of TP; however, the effect was, to some extent, enhanced with the presence of OβG. This might be because OβG induced a more expanded network of gluten, favoring the access and attack of TP to unfold or disrupt the gluten structure by breaking disulfide bonds, as confirmed by the red-shifts in fluorogram, increasing content of free sulfhydryl by 250 % (without OβG) and 312 % (with OβG), and decreasing particle size of gluten by 10.43 % (without OβG) and 21.08 % (with OβG) when the addition of TP was 2 %. Moreover, with the increasing of TP, the tremendous unfolding or disrupting gluten structure exposed more amino acids whereas decreased the intermolecular contacts and extended chains of gluten, consequently leading to the increasing hydrogen bonds and hydrophobic interactions while reducing the content of β-sheets, respectively.

A wavelet packet transform-based domain adaptation method for rolling bearings fault diagnosis with fusion of local and global time–frequency features

Abstract Fault diagnosis transfer learning models commonly employ deep neural networks (DNN) to analyze time-frequency features. However, excessively deep neural networks can result in diminished generalization capabilities, leading to subpar performance of the model across various working conditions. Furthermore, inappropriate domain adaptation strategies significantly constrain the accuracy of the model. To address this issue, a Robustly Optimized Residual-Network and Vision Transformer Domain Adaptation Model (RoReViTDAM) is proposed in this article, combining Wavelet Packet Transform (WPT), residual networks, and self-attention mechanisms. Firstly, the WPT is utilized to construct Multi-band Wavelet Coefficient Matrix (MWCM) and corresponding Multi-band Wavelet Coefficient Time-Frequency Feature Matrix (MWSM) with small size and feature aggregation. Subsequently, a shallow Robustly Optimized Residual Network (RoResNet) is designed to effectively extract features from MWCM, considering the spatial distance dependencies of features. Additionally, ViT (Vision Transformer) is employed for time-frequency global feature extraction from MWSM. Furthermore, Domain Adversarial Neural Network (DANN) and Multi-Kernel Maximum Mean Discrepancy (MK-MMD) are employed to extract domain-invariant features from signals of different operating conditions and fault types. At last, three fault diagnosis experiments are conducted in multi-condition scenarios of bearings. The experimental results illustrate the superiority and effectiveness of the proposed model.

Dual-domain feature aggregation transformer network for underwater image enhancement

Dual-domain feature aggregation transformer network for underwater image enhancement

Using Smartphone GPS Data to Detect the Risk of Adolescent Suicidal Thoughts and Behaviors.

Suicide rates among adolescents continue to rise, but there are a lack of clinical tools to predict when youths may be at risk for suicidal behaviors. To identify whether geolocation metrics, assessed through an app installed on adolescents' personal smartphones, could detect the risk of next-week suicidal events and clinically meaningful suicidal ideation. This case series study included high-risk adolescents aged 13 to 18 years reporting a current affective and/or substance use disorder, oversampled for suicidal thoughts and behaviors (STB). Participants were recruited from the greater New York City and Pittsburgh communities through psychiatric outpatient programs, emergency departments, medical center research registries, and social media. Participants installed the Effortless Assessment Research System (EARS) software application onto their personal smartphones, which obtained passive sensor data, including geolocation metrics (via the global positioning system [GPS]), as well as weekly experience sampling data probing STB for the duration of the 6-month study. Adolescents also completed clinical assessments at baseline as well as during the 1-, 3-, and 6-month follow-up assessments. Statistical analysis was performed from March 2023 to November 2024. Repeated measures mixed-effects logistic models estimated whether weekly aggregates of geolocation features (ie, entropy, homestay, distance traveled) were associated with next-week suicidal events (ie, suicide attempts, psychiatric hospitalizations, emergency department visits for suicide concerns) and clinically meaningful ideation (via weekly experience sampling). Overall, 186 participants were included in this study (148 [79.6%] female; 19 [10.2%] Asian, 23 [12.4%] Black, and 106 [57.0%] White), with a mean (SD) age of 16.4 (1.7) years. Greater homestay (amount of time spent at home) on a given week, relative to one's own mean, was associated with 2-fold greater odds of suicidal events during the subsequent week (odds ratio, 1.99 [95% CI, 1.15-3.45]). Results were not significant for entropy and distance traveled metrics. However, using leave-future-out validation, the accuracy of the homestay model was modest (area under the receiver operating characteristic curve, 0.64 [95% CI, 0.50-0.78]). Advancements in smartphone technology afford unique opportunities to capture affective and behavioral dynamics that presage suicide risk. This case series study found that greater homestay obtained through smartphone GPS data over the course of a week, relative to one's own mean, was associated with greater odds of a suicidal event in the subsequent week. Although accuracy was modest, these findings offer a novel starting point for suicide prevention research, particularly as smartphone sensor data may have the capacity to identify who is at risk while also providing an opportunity to deliver clinical tools when that risk is greatest.

Composition and liquid-to-solid maturation of protein aggregates contribute to bacterial dormancy development and recovery

Recalcitrant bacterial infections can be caused by various types of dormant bacteria, including persisters and viable but nonculturable (VBNC) cells. Despite their clinical importance, we know fairly little about bacterial dormancy development and recovery. Previously, we established a correlation between protein aggregation and dormancy in Escherichia coli. Here, we present further support for a direct relationship between both. Our experiments demonstrate that aggregates progressively sequester proteins involved in energy production, thereby likely causing ATP depletion and dormancy. Furthermore, we demonstrate that structural features of protein aggregates determine the cell’s ability to exit dormancy and resume growth. Proteins were shown to first assemble in liquid-like condensates that solidify over time. This liquid-to-solid phase transition impedes aggregate dissolution, thereby preventing growth resumption. Our data support a model in which aggregate structure, rather than cellular activity, marks the transition from the persister to the VBNC state.

A Coarse-to-Fine Feature Aggregation Neural Network with a Boundary-Aware Module for Accurate Food Recognition

Food recognition from images is crucial for dietary management, enabling applications like automated meal tracking and personalized nutrition planning. However, challenges such as background noise disrupting intra-class consistency, inter-class distinction, and domain shifts due to variations in capture angles, lighting, and image resolution persist. This study proposes a multi-stage convolutional neural network-based framework incorporating a boundary-aware module (BAM) for boundary region perception, deformable ROI pooling (DRP) for spatial feature refinement, a transformer encoder for capturing global contextual relationships, and a NetRVLAD module for robust feature aggregation. The framework achieved state-of-the-art performance on three benchmark datasets, with Top-1 accuracies of 99.80% on the Food-5k dataset, 99.17% on the Food-101 dataset, and 85.87% on the Food-2k dataset, significantly outperforming existing methods. This framework holds promise as a foundational tool for intelligent dietary management, offering robust and accurate solutions for real-world applications.

Edge-aware Feature Aggregation Network for Polyp Segmentation

Edge-aware Feature Aggregation Network for Polyp Segmentation

Multi-task interaction learning for accurate segmentation and classification of breast tumors in ultrasound images.

In breast diagnostic imaging, the morphological variability of breast tumors and the inherent ambiguity of ultrasound images pose significant challenges. Moreover, multi-task computer-aided diagnosis systems in breast imaging may overlook inherent relationships between pixel-wise segmentation and categorical classification tasks. 
Approach. In this paper, we propose a multi-task learning network with deep inter-task interactions that exploits the inherently relations between two tasks. First, we fuse self-task attention and cross-task attention mechanisms to explore the two types of interaction information, location and semantic, between tasks. In addition, a feature aggregation block is developed based on the channel attention mechanism, which reduces the semantic differences between the decoder and the encoder. To exploit inter-task further, our network uses an circle training strategy to refine heterogeneous feature with the help of segmentation maps obtained from previous training. 
Main results. The experimental results show that our method achieved excellent performance on the BUSI and BUS-B datasets, with DSCs of 81.95% and 86.41% for segmentation tasks, and F1 scores of 82.13% and 69.01% for classification tasks, respectively. 
Significance. The proposed multi-task interaction learning not only enhances the performance of all tasks related to breast tumor segmentation and classification but also promotes research in multi-task learning, providing further insights for clinical applications.}.

Few-Shot Object Detection for SAR Images via Context-Aware and Robust Gaussian Flow Representation

In recent decades, few-shot object detection in SAR imagery has gained prominence as a major research focus. The unique imaging mechanism of SAR causes the model to suffer from foreground–background imbalance and inaccurate extraction of class prototypes for novel class instances. Therefore, we propose an innovative few-shot object detection algorithm for SAR images via context-aware and robust Gaussian flow representation. First, we design the Context-Aware Enhancement module to address the foreground–context imbalance problem by refining representative support features into fine-grained prototypes, which are deeply fused with query features based on the prototype matching paradigm. Second, we devise the Manifold Class Distribution Estimation module to address the difficulty of class distribution estimation and the fluctuation of class centers of the sparse novel class. Furthermore, we formulate the Category-Balanced Difference Aggregation module to model the relationship between the base class and the novel class, addressing the sensitivity of the model to the variance of the novel class instances. Finally, we design the Cosine Decoupling Module so that the aggregated features are executed only on the classification branch without affecting the precise localization of the target. Experiments based on SAR-AIRcraft-1.0 and the self-constructed MSAR-AIR dataset indicate that the fine-grained detection and identification performance of the novel class of airplanes can reach 32.90% and 55.26%, respectively, in the 10-shot and 50-shot cases. In addition, our method enables cross-domain detection for different scenarios and sample types and exhibits excellent generalization performance in data-sparse scenarios.

BurgsVO: Burgs-Associated Vertex Offset Encoding Scheme for Detecting Rotated Ships in SAR Images

Synthetic Aperture Radar (SAR) is a crucial remote sensing technology with significant advantages. Ship detection in SAR imagery has garnered significant attention. However, existing ship detection methods often overlook feature extraction, and the unique imaging mechanisms of SAR images hinder the direct application of conventional natural image feature extraction techniques. Moreover, oriented bounding box-based detection methods often prioritize accuracy excessively, leading to increased parameters and computational costs, which in turn elevate computational load and model complexity. To address these issues, we propose a novel two-stage detector, Burgs-rooted vertex offset encoding scheme (BurgsVO), for detecting rotated ships in SAR images. BurgsVO consists of two key modules: the Burgs equation heuristics module, which facilitates feature extraction, and the average diagonal vertex offset (ADVO) encoding scheme, which significantly reduces computational costs. Specifically, the Burgs equation module integrates temporal information with spatial data for effective feature aggregation, establishing a strong foundation for subsequent object detection. The ADVO encoding scheme reduces parameters through anchor transformation, leveraging geometric similarities between quadrilaterals and triangles to further reduce computational costs. Experimental results on the RSSDD and RSDD benchmarks demonstrate that the proposed BurgsVO outperforms the state-of-the-art detectors in both accuracy and efficiency.

Computational spectral imaging reconstruction via a spatial–spectral cross-attention-driven network

Compared with traditional hyperspectral imaging, computational spectral imaging (CSI) has the advantage of snapshot imaging with high spatial and temporal resolution, which has attracted considerable attention. The core challenge of CSI is to achieve computational imaging reconstruction from a single 2D measurement image to the corresponding 3D spatial–hyperspectral image (HSI). Existing reconstruction methods still face problems in exploring spatial–spectral cross correlation, leading to significant spatial–spectral distortion. Furthermore, due to neglect of multi-scale feature reconstruction, their reconstruction quality still needs to be improved. In this paper, to solve the above problems, we propose a spatial–spectral cross-attention-driven network (SSCA-DN). In SSCA, a proposed multi-scale feature aggregation (MFA) module and a spectral-wise transformer (SpeT) are used for multi-scale spatial feature reconstruction and long-range spectral feature reconstruction, respectively. Using spatial attention and spectral attention to interactively guide the reconstruction of the target HSI in spectral and spatial dimensions, the proposed SSCA models spatial–spectral cross correlation with considering multi-scale features. Using the SSCA as a basic module, a novel SSCA-DN network is constructed, in which a proposed supervised preliminary reconstruction subnetwork (SPRNet) learns the generalized prior, and a proposed unsupervised multi-scale feature fusion and refinement subnetwork (UMFFRNet) learns the specific prior. The SSCA module ensures that the learned generalized and specific priors can capture the spatial–spectral cross correlation while considering multi-scale features. In addition, in UMFFRNet, driven by MFA and SSCA, a novel multi-scale fusion and refinement mechanism for multi-level adjacent features is proposed to effectively model the correlation between adjacent level features and the multi-scale spatial–spectral cross correlation, which further improves the reconstruction accuracy. Extensive experiments show that our method achieves state-of-the-art performance on both simulated and real datasets.

Green Recycled Aggregate in Concrete: Feasibility Study

With increasing concrete production, CO2 emissions rise, and natural resources deplete, creating a need for new material solutions. This article analyzes the feasibility of using green materials, like recycled aggregate (RA) from construction and demolition waste (CDW) to be incorporated into concrete (RAC). The objective of this paper is to determine that the use of RA ensures receiving sustainable concrete in comparison with NA and LA. The sustainability assessment was conducted based on an analysis of the life cycle in terms of the environmental, economic, and public perception aspects. Additionally, the analysis was extended to include two newly introduced indicators: quality of aggregates and concrete performance. A proprietary scoring method based on ideal aggregate characteristics was used, which was enhanced by innovative multidimensional analysis, with credits assigned based on a literature review conducted using artificial intelligence (AI) statistical tools to partially assist in the selection of items. The results could even show that RA outperformed natural aggregates (NA) and artificial (light) aggregates (LA) in the environmental (over 80% of the results) results as well as the economic (over 65%) and public perception categories (over 80%). However, RA ranked second behind NA in terms of quality aggregates and concrete performance, with LA scoring lowest. The results highlight RAC as a satisfactory sustainable option compared with NAC, supporting the circular economy by reducing waste, emissions, and resource consumption. The best solution would be hybrid concrete containing a partial substitute for natural aggregates in the form of recycled aggregates, enabling the advantages of both types of aggregates to complement each other and offset their limitations.

Photosensitizable ZIF-8 BioMOF for Stimuli-Responsive Antimicrobial Phototherapy.

Resistant pathogens are increasingly posing a heightened risk to healthcare systems, leading to a growing concern due to the lack of effective antimicrobial treatments. This has prompted the adoption of antimicrobial photodynamic therapy (aPDT), which eradicates microorganisms by generating reactive oxygen species (ROS) through the utilization of a photosensitizer, photons, and molecular oxygen. However, a challenge arises from the inherent characteristics of photosensitizers, including photobleaching, aggregation, and self-quenching. Consequently, a strategy has been devised to adsorb or bind photosensitizers to diverse carriers to facilitate their delivery. Notably, metal-organic frameworks (MOFs) have emerged as a promising means of transporting photosensitizers, even though achieving uniform particle sizes through room-temperature synthesis remains a complex task. In this work, we have tackled the issue of heterogeneous particle size distribution in MOFs, achieving a particle size of 150 ± 50 nm. Subsequently, we harnessed Zeolite Imidazolate Framework 8 (ZIF-8), an excellent subclass of biocompatible MOF, to effectively load two distinct categories of photosensitizers, namely, Rose Bengal (RB) and porphyrin, using a simple, straightforward, and single-step process. Our findings indicate that the prepared RB@ZIF-8 complex generates a more substantial amount of reactive singlet oxygen species when subjected to photoirradiation (using green light-emitting diode (LED)) at low concentrations, in comparison with porphyrin@ZIF-8, as demonstrated in in vitro experiments. Additionally, we investigated the pH-responsive behavior of the complex to ascertain its implications under biological conditions. Correspondingly, the RB@ZIF-8 complex exhibited a more favorable IC50 value against Escherichia coli compared to bare photosensitizers, ZIF-8 alone, and other photosensitizer-loaded ZIF-8 complexes. This underscores the potential of BioMOF as a promising strategy for combatting multidrug-resistant bacteria across a spectrum of infection scenarios, complemented by its responsiveness to stimuli.

Nonlinear Correct and Smooth for Graph-Based Semi-Supervised Learning

Graph-based semi-supervised learning (GSSL) has achieved significant success across various applications by leveraging the graph structure and labeled samples for classification tasks. In the field of GSSL, Label Propagation (LP) and Graph Neural Networks (GNNs) are two complementary methods, in which LP iteratively propagates and updates node labels through connected nodes, whereas GNNs aggregate node features by incorporating information from their neighbors. Recently, the complementary nature of LP and GNNs has been utilized to improve performance through the combination of two approaches. However, the utilization of higher-order graph structures within these combined approaches, such as triangles, is still underexplored. Therefore, to advance understanding in this ongoing research, we first model GSSL as a two-step Feature-Label process. Then, we introduce Nonlinear Correct and Smooth (NLCS) in the post-processing step, a combined method that incorporates nonlinearity and higher-order structures into the residual propagation to handle intricate node relationships effectively. We propose a new synthetic graph generator to deepen the analysis and broaden the experimentation, providing insights into the mechanisms that enable NLCS to handle intricate node relationships effectively. Our systematic evaluations across six synthetic graphs show that NLCS outperforms base predictions by an average of 12.44% and the existing state-of-the-art post-processing method by 8.04%. Furthermore, on six commonly used real-world datasets, NLCS demonstrates a 10.9% improvement over six base prediction models and a 1.6% over the state-of-the-art post-processing method. Our comparisons and analyses reveal that NLCS substantially enhances the prediction accuracy of nodes within complex graph structures by effectively utilizing higher-order structures of graphs.