Network Modules Research Articles

Research on Small-Target Detection of Flax Pests and Diseases in Natural Environment by Integrating Similarity-Aware Activation Module and Bidirectional Feature Pyramid Network Module Features

In the detection of the pests and diseases of flax, early wilt disease is elusive, yellow leaf disease symptoms are easily confusing, and pest detection is hampered by issues such as diversity in species, difficulty in detection, and technological bottlenecks, posing significant challenges to detection efforts. To address these issues, this paper proposes a flax pest and disease detection method based on an improved YOLOv8n model. To enhance the detection accuracy and generalization capability of the model, this paper first employs the Albumentations library for data augmentation, which strengthens the model’s adaptability to complex environments by enriching the diversity of training samples. Secondly, in terms of model architecture, a Bidirectional Feature Pyramid Network (BiFPN) module is introduced to replace the original feature extraction network. Through bidirectional multi-scale feature fusion, the model’s ability to distinguish pests and diseases with similar features and large scale differences is effectively improved. Meanwhile, the integration of the SimAM attention mechanism enables the model to learn information from three-dimensional channels, enhancing its perception of pest and disease features. Additionally, this paper adopts the EIOU loss function to further optimize the model’s bounding box regression, reducing the distortion of bounding boxes caused by high sample variability. The experimental results demonstrate that the improved model achieves a significant detection performance on the flax pest and disease dataset, with notable improvements in the detection accuracy and mean average precision compared to the original YOLOv8n model. Finally, this paper proposes a YOLOv8n model with a four-headed detection design, which significantly enhances the detection capability for small targets such as pests and diseases with a size of 4 × 4 pixels or larger by introducing new detection heads and optimizing feature extraction. This method not only improves the detection accuracy for flax pests and diseases but also maintains a high computational efficiency, providing effective technical support for the rapid and precise detection of flax pests and diseases and possessing an important practical application value.

Microbial generalists as keystone species: constructing core network modules in the anthosphere of twelve diverse wild plant species

BackgroundThe anthosphere, also known as the floral microbiome, is a crucial component of the plant reproductive system. Therefore, understanding the anthospheric microbiome is essential to explore the diversity, interactions, and functions of wildflowers that coexist in natural habitats. We aimed to explore microbial interaction mechanisms and key drivers of microbial community structures using 144 flower samples from 12 different wild plant species inhabiting the same natural environment in South Korea.ResultsThe microbial diversity of the anthosphere showed plant dependence, with the highest diversity observed in Forsythia koreana, indicating microbial dynamics in relation to plant species. Caulobacter, Sphingomonas, Achromobacter, Epicoccum, Cladosporium, and Alternaria were anthosphere generalists, suggesting that the local plant anthosphere had a similar microbial composition. Ecological network analysis revealed that anthosphere generalists were tightly coupled to each other and constructed core modules in the anthosphere. Functions associated with parasites and pathogens were commonly observed in the anthosphere, particularly in Capsella bursa-pastoris and Brassica juncea.ConclusionOverall, the anthosphere depends on the plant species and microbial generalists function as keystone species to support and connect the anthospheric microbiome in natural habitats.

Networks of pre-diagnostic circulating RNA in testicular germ cell tumour

Testicular germ cell tumour (TGCT) is a malignancy with known inherited risk factors, affecting young men. We have previously identified several hundred differentially abundant circulating RNAs in pre-diagnostic serum from TGCT cases compared to healthy controls. In this study, we performed Weighted Gene Co-expression Network Analysis (WGCNA) on mRNA and miRNA data from these samples. Central genes (hub genes) enriched functional pathways, and regulatory feature prediction were identified for all TGCT subtypes together and according to histology. The TGCT susceptibility genes TEX14, NARS2, and G3BP2, were identified as hub genes in both seminoma and non-seminoma networks. We also identified UBCA1, RCC1, FMR1, OAS3, and UBE2W as hub genes associated with TGCT. The genes OAS3 and UBE2W have previously been associated with testicular dysgenesis. Furthermore, network module analysis indicated transcription factors for oestrogen-related receptors to have a potential role during development of TGCT. The overlap between mRNA network hub genes and TGCT susceptibility genes indicates a common role in TGCT development.

Network-Based Identification of Key Toxic Compounds in Airborne Chemical Exposome.

Air pollution is a leading contributor to the global disease burden. However, the complex nature of the chemicals to which humans are exposed through inhalation has obscured the identification of the key compounds responsible for diseases. Here, we develop a network topology-based framework to identify key toxic compounds in the airborne chemical exposome. Using cardiovascular diseases (CVDs) as a model disease, we found that toxic network modules of various compounds are closely linked to the modules of CVDs. The proximity of compound target modules to disease modules can indicate the extent of toxicity induced by the compounds. By integrating mass spectrometry-based external exposure concentrations and machine learning-predicted internal exposure concentrations, we established a comprehensive linkage connecting exposure to disease-related risk for the identification of toxic compounds. These findings were subsequently validated using exposure and disease data on the regional scale. This work provides an effective strategy for identifying key compounds within environmental exposomes and establishes a new paradigm for understanding the pathogenicity of air pollution.

Motile cilia modulate neuronal and astroglial activity in the zebrafish larval brain.

The brain uses a specialized system to transport cerebrospinal fluid (CSF), consisting of interconnected ventricles lined by motile ciliated ependymal cells. These cells act jointly with CSF secretion and cardiac pressure gradients to regulate CSF dynamics. To date, the link between cilia-mediated CSF flow and brain function is poorly understood. Using zebrafish larvae as a model system, we identify that loss of ciliary motility does not alter progenitor proliferation, brain morphology, or spontaneous neural activity despite leading to an enlarged telencephalic ventricle. We observe altered neuronal responses to photic stimulations in the optic tectum and hindbrain and brain asymmetry defects in the habenula. Finally, we investigate astroglia since they contact CSF and regulate neuronal activity. Our analyses reveal a reduction in astroglial calcium signals during both spontaneous and light-evoked activity. Our findings highlight a role of motile cilia in regulating brain physiology through the modulation of neural and astroglial networks.

MicroRNA analysis reveals two modules that antagonistically regulate xylem tracheary element development in Arabidopsis.

Tracheary elements (TEs) are vital in the transport of various substances and contribute to plant growth. The differentiation of TEs is complex and regulated by a variety of microRNAs (miRNAs). However, the dynamic changes in miRNAs during each stage of TE differentiation remain unclear, and the miRNA regulatory network is not yet complete. This study employed Vascular cell Induction culture System Using Arabidopsis Leaves (VISUAL) to profile the miRNome during TE differentiation in Arabidopsis (Arabidopsis thaliana) and established comprehensive miRNA co-expression networks functioning at the different stages of TE differentiation. Two negatively correlated modules exist in the miRNA networks, each exhibiting strong intra-module positive correlation and strong inter-module negative correlation. Thus, the two modules may play opposite roles in TE differentiation and vascular development. Indeed, we found that miR408 promotes cambium formation and TE differentiation, consistent with miR408 as a key node in the networks of fate determination and the initiation of TE differentiation. Additionally, we found that miR163 inhibits secondary cell wall formation and TE differentiation, corresponding to miR163 as a key node in the TE maturation network. Moreover, we discovered that the miRNA co-expression network in poplar (Populus tomentosa) xylem development is also composed of two negatively correlated modules that contain miRNAs orthologous to those in Arabidopsis. Therefore, the two negatively correlated modules of the miRNA co-expression network are likely conserved and fundamental to xylem TE differentiation. These results provide insights into microRNA regulation in plant development.

Concrete sleeper crack detection based on image saliency

Abstract Concrete sleepers are crucial infrastructure in railway construction and operation. Sleeper cracks may cause structural damage or functional failure of sleepers, thereby endangering railway operation safety. A third-order residual crack saliency detection network (R3Net CSD) model was constructed based on ConvNeXt for detecting concrete sleeper cracks in track inspection images, which is used to accurately extract the saliency features of rail sleeper crack images. The model first uses ConNeXt to form the backbone network and Block module to form the first residual, followed by a spatial pyramid (DCSP) module based on the anti residual structure and dilated convolution group to form the second residual, and finally uses the saliency residual (Att-R) module to form the third residual. At the same time, combined with a multi-level joint loss function supervision strategy for supervised learning, the model fully utilizes the low-level edge information and high-level semantic information of cracks. Using the images collected by the hand pushed rail inspection vehicle, a dataset of rail sleeper cracks was manually constructed and the model was experimentally validated. The experimental results show that this model achieved the best results in multiple indicators, such as an average absolute error MAE of 0.0082, an average accuracy AP of 0.9764, and an AUC value of 0.995. By combining the calculation method of crack geometric parameters, quantitative indicators of crack length and width were obtained, with a detection accuracy of 0.001mm. The research provides a new solution for the detection of cracks in concrete sleepers and the measurement of geometric parameters.

The Impact of PTEN Status on Glioblastoma Multiforme: A Glial Cell Type-Specific Study Identifies Unique Prognostic Markers

Glioblastoma multiforme (GBM) is the most invasive form of brain tumor, accounting for 5% of the cases per 100,000 people in various countries. The phosphatase and tensin homolog deleted from chromosome 10 (PTEN) is a well-known tumor suppressor, and its alteration leads to a deleterious effect on GBM progression. The molecular mechanism of tumorigenesis in glial cell types, driven by PTEN status, is yet to be elucidated. In this study, we analyzed publicly available single-cell transcriptome profiles of PTEN wild-type (WT) and NULL GBM patients. We compared them with normal brain data to uncover many unique gene sets influenced by PTEN status. The co-expression network analysis of differentially expressed genes (DEGs) between normal brain and PTEN (WT and NULL) identified highly interconnected genes. The weighted gene co-expression network analysis (WGCNA), based on the DESeq2 algorithm, identified glial cell-type-specific modules in PTEN status-dependent bulk RNA expression profiles. We overlapped network module gene sets from single-cell and bulk transcriptome profiles, and shared genes were considered for further analysis. The hallmark pathway enrichment analysis of the genes unique to PTEN-WT and NULL revealed various tumor growth-related pathways across the glial cell types. Further characterization of PTEN-WT and PTEN-NULL networks belonging to the single-cell and bulk RNA datasets revealed that PTEN status influences the network modules in astrocytes, microglia, and oligodendrocyte precursor cells. An integrated influence value algorithm identified hub genes for each glial cell type. The prognostic analysis identified clinically relevant hub genes specific to the cell type in PTEN-WT: GLIPR2 (astrocytes), CFH, IL32, MXRA5 (microglia), and PTEN-NULL: ID1 (astrocytes) and LAT2 (microglia). Our glial cell type-level transcriptome analysis unearthed unique molecular pathways and prognostic markers in PTEN status-dependent GBM patients.

A knowledge-refined hybrid graph model for quality prediction of industrial processes

The complexity of industrial processes has spurred the application of soft sensor techniques for predicting key quality variables based on easy-measurable process variables. Currently, data-driven soft sensors based on Artificial Intelligence techniques have become the mainstream. However, these soft sensing models deeply rely on the quality of training data, where the domain knowledge is often ignored. Meanwhile, a significant amount of labeled data is not fully utilized. To address these issues, this paper proposes a supervised framework based on a knowledge-refined hybrid graph network, which contributes to the artificial intelligence application of nonlinear dynamic soft sensors. The problems of applying traditional artificial intelligence models in soft sensor have been addressed by reconstructing the input module of graph neural networks with knowledge-guided approaches. Both spatial and temporal correlations of process data are captured and the hybrid network significantly improves the reliability and interpretability of the soft sensing model. By incorporating labeled data into the model, the representation of quality information is also enhanced. Finally, the proposed framework was applied to an industrial debutanizer column, and the experimental results fully demonstrate the effectiveness and superiority of the method.

Multimodal neurophenomenology of advanced concentration absorption meditation: An intensively sampled case study of Jhana.

Using a combination of fMRI, EEG, and phenomenology ratings, we examined the neurophenomenology of advanced concentrative absorption meditation, namely jhanas (ACAM-J), in a practitioner with over 23,000 h of meditation practice. Our study shows that ACAM-J states induce reliable changes in conscious experience and that these experiences are related to neural activity. Using resting-state fMRI functional connectivity, we found that ACAM-J is associated with decreased within-network modularity, increased global functional connectivity (GFC), and desegregation of the default mode and visual networks. Compared to control tasks, the ACAM-J were also related to widespread decreases in broadband EEG oscillatory power and increases in Lempel-Ziv complexity (LZ, a measure of brain entropy). Some fMRI findings varied by the control task used, while EEG results remained consistent, emphasizing both shared and unique neural features of ACAM-J. These differences in fMRI and EEG-measured neurophysiological properties correlated with specific changes in phenomenology - and especially with ACAM-J-induced states of bliss - enriching our understanding of these advanced meditative states. Our results show that advanced meditation practices markedly dysregulate high-level brain systems via practices of enhanced attention to sensations, corroborating recent neurocognitive theories of meditation as the deconstruction of the brain's cortical hierarchy. Overall, our results suggest that ACAM-J is associated with the modulation of large-scale brain networks in both fMRI and EEG, with potential implications for understanding the mechanisms of deep concentration practices and their effects on subjective experience.

Mechanism of Compound Kushen Injection in the Treatment of Acute Myeloid Leukemia from the Analysis Perspectives

Background: Chemotherapy resistance often occurs in the conventional treatment with AML and results in poor cure rates. CKI was found to have a good therapeutic effect when it was combined with other chemotherapy drugs in the clinical treatment of AML. However, the underlying mechanism is unclear. Therefore, this study aims to preliminarily describe the pharmacological activity and mechanism of CKI through comprehensive network pharmacology methods. Objective: This study aimed to explore the possible mechanism of Compound Kushen Injection (CKI) in the treatment of acute myeloid leukemia (AML) by using network pharmacology, molecular docking, and molecular dynamics techniques. Methods: Active compounds of CKI were identified based on the Traditional Chinese Medicine Systems Pharmacy (TCMSP) database, and the related targets of the active compounds were predicted using Swiss Target Prediction; AML-related targets from Gene Cards and Online Mendelian Inheritance in Man (OMIM) were collected. Protein-protein interaction (PPI) network was constructed, and its mechanism was predicted through Gene Ontology (GO) analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment. The protein-protein interaction (PPI) network construction, module partitioning, and hub node screening were visualized by using the Cytoscape software and its plugins. These module partitionings were also verified by using molecular docking and molecular dynamics modeling. Results: Fifty-six active ingredients corresponding to 223 potential targets were identified. Biological function analysis showed that 731, 70, and 137 GO entries were associated with biological processes, cellular components, and molecular functions, respectively. A total of 163 KEGG pathways were identified. Network analysis showed that the key anti-AML targets of CKI are MAPK3, EGFR, SRC, PIK3CA, and PIK3R1 targets, which are involved in the PI3K/Akt and Ras/MAPK signaling pathways or related crosstalk pathways. Conclusion: Our results suggested that the key anti-AML targets of CKI, such as MAPK3, EGFR, SRC, PIK3CA and PIK3R1, are involved in the PI3K/Akt and Ras/MAPK signaling pathways or related crosstalk pathways. Concentrating on the dynamic and complex crosstalk regulation between PI3K/Akt and Ras/MAPK signal pathways and related signal pathways may be a new direction in anti-AML therapy in the future.

Comprehensive analyses reveal the promising value of gasdermins as prognostic biomarkers and immunotherapeutic targets in head and neck squamous cell carcinoma.

In several studies of head and neck squamous cell carcinoma (HNSC), the regulation of tumorigenesis and therapeutic sensitivity by pyroptosis has been observed. However, a systematic analysis of gasdermin family members (GSDMs, including GSDMA/B/C/D/E and PJVK), which are deterministic executors of pyroptosis, has not yet been reported in HNSC. We performed comprehensive analyses of the expression profile, prognostic value, regulatory network, and immune infiltration modulation of GSDMs in HNSC on the basis of a computational approach and bioinformatic analysis of publicly available datasets. A total of 18.65% (94/504) of HNSC patients harbored GSDM alterations, with the most dominant type being amplification. Compared with those in normal tissues, the mRNA and protein levels of GSDMs, especially GSDMD/E, were commonly elevated in HNSC (P<0.05). Additionally, the expression of GSDMs differed significantly between the clinicopathological subgroups of HNSC patients. Overall survival of HNSC patients benefited from increased GSDMC expression (HR=0.67, P=0.0053) and decreased GSDME expression (HR=1.42, P=0.0140). Regulatory network analysis revealed several essential biological processes associated with GSDMs, including positive regulation of cytokine production involved in the immune response. Notably, almost all infiltrating immune cells and immune checkpoints were negatively correlated with GSDMA/C/E expression and positively related to GSDMB/D and PJVK expression. We indicated the potential role of GSDMs (especially GSDME) in HNSC pathogenesis, progression and response to immunotherapy, providing important evidence for further prospective studies and molecular mechanism exploration.

A deep learning‐based method for detecting and identifying surface defects in polyimide foam

AbstractCurrently, the detection and identification of surface defects in polyimide foam products mainly rely on on‐site work experience, which has issues such as low detection accuracy, strong subjectivity, and low efficiency. Existing research on foam product defect detection primarily targets internal defects, lacking studies on the detection, identification, and classification of surface defects. Therefore, this article proposes a method for identifying and classifying surface defects in polyimide foam based on an improved GoogLeNet, aiming to quickly and accurately detect and identify surface defects in foam products. By optimizing the Inception blocks, introducing the ECA attention mechanism, and adding an LSTM network module, the model's recognition accuracy and generalization ability are effectively improved. In experiments, the model proposed in this article performed excellently on the foam surface defect dataset, showing a significant advantage in detection accuracy compared to other convolutional neural network models. The detection accuracy for pits and cracks reached 98.24% and 98.25%, respectively, providing a reliable reference for the detection of surface defects in industrial foam production.

CT-RURnet: a novel network design for radar unmanned aerial vehicles recognition

Abstract The rapid growth of commercial unmanned aerial vehicles (UAVs) has led to more incidents. These include illegal activities by malicious actors. Such activities pose significant threats to public facilities and people’s safety. Therefore, distinguishing UAVs from other targets is a critical measure in preventing potential hazards. This paper proposes an enhanced feature extraction and global modeling model for UAV recognition, named CT-RURnet. The model directly learns from radar echo data, firstly using the convolutional neural networks module to quickly extract local features, and then using the Transformer module to establish global connections among these local features. The proposed model primarily focuses on learning the micro-Doppler features produced by targets, which facilitates UAV recognition. The used datasets include both simulated data and the Real Doppler RAD-DAR (RDRD) dataset. The proposed network achieves accuracy rates of 98.7% (10 SNR), 97.9% (5 SNR), and 86.95% (0 SNR) on the simulated dataset under varying SNR conditions and 97.14% on the real dataset. Compared to other baseline models, the CT-RURnet consistently delivers superior results. Ablation experiments are conducted to verify the contribution of each module to the overall network performance.

Enhancement of Composite Material Properties Based on Aggregate Network Modulation

ABSTRACTCompounding technology and the theory of filled polymers have progressed substantially. Valuable insight from these developments is that carbon black‐filled wet rubber greatly enhances the filler's dispersion and dispersion state in the matrix. The enhanced dispersion noticeably improves the properties of the polymer. This paper systematically investigated the effects of silanization reactions on filler dispersion, interaction force, and aggregation networks. Their potential impact on polymers' properties was also evaluated. To this end, we evaluated the effects of different ratios of Si69 and silica fillers in wet and dry rubbers. For better characterization and utilization of the filler network structure, the filler–filler, filler–rubber, and rubber–rubber interactions were analyzed using the Mooney–Rivlin curve model. The results revealed that wet blending and silanization reactions primarily regulated the aggregate size and filler network strength. Increasing silica content improved the Payne effect. Notably, incorporating an appropriate filler content could effectively enhance rubber's physical and mechanical properties. This paper effectively optimized the structure and properties of filled polymers by controlling silanization reaction and applying wet mixing. This research contributes to the theoretical and experimental foundation of studies in relevant fields.

TMS-induced modulation of brain networks and its associations to rTMS treatment for depression: a concurrent fMRI-EEG-TMS study.

Transcranial magnetic stimulation (TMS) over the left dorsolateral prefrontal cortex (L-DLPFC) is an established intervention for treatment-resistant depression (TRD), yet the underlying therapeutic mechanisms remain not fully understood. This study employs an integrative approach that combines TMS with concurrent functional magnetic resonance imaging (fMRI) and electroencephalography (EEG), aimed at assessing the acute/immediate effects of TMS on brain network dynamics and their correlation with clinical outcomes. Our study demonstrates that TMS acutely modulates connectivity within vital brain circuits, particularly the cognitive control and default mode networks. We found that the baseline TMS-evoked responses in the cognitive control and limbic networks significantly predicted clinical improvement in patients receiving a novel EEG-synchronized repetitive TMS treatment. Furthermore, this study explored the brain-state dependent effects of TMS, as the brain-state indexed by the phase of EEG prefrontal alpha oscillation. We found that clinical outcomes in this novel treatment are linked to state-specific TMS-modulated functional connectivity within a pivotal brain circuit of the L-DLPFC and the posterior subgenual anterior cingulate cortex within the limbic system. These findings contribute to our understanding of the therapeutic effects underlying TMS treatment in depression and support the potential of assessing state-dependent TMS effects in TMS timing target selection. This study emphasizes the importance of personalized timing of TMS for optimizing target engagement of specific clinically relevant brain circuits. Our results are crucial for future research into the development of personalized neuromodulation therapies for TRD patients.

The development of a waste management and classification system based on deep learning and Internet of Things.

Waste sorting is a key part of sustainable development. To maximize the recovery of resources and reduce labor costs, a waste management and classification system is established. In the system, we use Internet of Things (IoT) and edge computing to implement waste sorting and the systematic long-distance information transmission and monitoring. A dataset of recyclable waste images with realistic backgrounds was collected, where the images contained multiple waste categories in a single image. An improved deep learning model based on YOLOv7-tiny is proposed to adapt to the realistic complex background of waste images. In the model, adding partial convolution (PConv) to Efficient Layer Aggregation Network (ELAN) module reduces parameters and floating point of operations (FLOPs). Coordinate attention (CA) is added to spatial pyramid pooling (Sppcspc) module and ELAN module, respectively. SIoU loss function is used, which improves the recognition accuracy of the model. The improved model shows a higher accuracy on the basis of lighter weight and is more suitable for deployment on edge devices. The proposed model and the original model were trained using our dataset, and their performance was compared. According to the experimental results, mAP@.5, mAP@.5:.95 of the improved YOLOv7-tiny are increased by 1.7% and 1.4%, and the parameter and FLOPs are decreased by 4.8% and 5%, respectively. The improved model has an average inference time of 110ms and an FPS of 9 on the Jetson Nano. Hence, we believe that the developed system can better meet the needs of current garbage collection system.

MiRNA target enrichment analysis of co-expression network modules reveals important miRNAs and their roles in breast cancer progression

Abstract Breast cancer has the highest incidence and is the fifth cause of death in cancers. Progression is one of the important features of breast cancer which makes it a life-threatening cancer. MicroRNAs are small RNA molecules that have pivotal roles in the regulation of gene expression and they control different properties in breast cancer such as progression. Recently, systems biology offers novel approaches to study complicated biological systems like miRNAs to find their regulatory roles. One of these approaches is analysis of weighted co-expression network in which genes with similar expression patterns are considered as a single module. Because the genes in one module have similar expression, it is rational to think the same regulatory elements such as miRNAs control their expression. Herein, we use WGCNA to find important modules related to breast cancer progression and use hypergeometric test to perform miRNA target enrichment analysis and find important miRNAs. Also, we use negative correlation between miRNA expression and modules as the second filter to ensure choosing the right candidate miRNAs regarding to important modules. We found hsa-mir-23b, hsa-let-7b and hsa-mir-30a are important miRNAs in breast cancer and also investigated their roles in breast cancer progression.

Detecting Ocean Eddies with a Lightweight and Efficient Convolutional Network

As a ubiquitous mesoscale phenomenon, ocean eddies significantly impact ocean energy and mass exchange. Detecting these eddies accurately and efficiently has become a research focus in ocean remote sensing. Many traditional detection methods, rooted in physical principles, often encounter challenges in practical applications due to their complex parameter settings, while effective, deep learning models can be limited by the high computational demands of their extensive parameters. Therefore, this paper proposes a new approach to eddy detection based on the altimeter data, the Ghost Attention Deeplab Network (GAD-Net), which is a lightweight and efficient semantic segmentation model designed to address these issues. The encoder of GAD-Net consists of a lightweight ECA+GhostNet and an Atrous Spatial Pyramid Pooling (ASPP) module. And the decoder integrates an Efficient Attention Network (EAN) module and an Efficient Ghost Feature Integration (EGFI) module. Experimental results show that GAD-Net outperforms other models in evaluation indices, with a lighter model size and lower computational complexity. It also outperforms other segmentation models in actual detection results in different sea areas. Furthermore, GAD-Net achieves detection results comparable to the Py-Eddy-Tracker (PET) method with a smaller eddy radius and a faster detection speed. The model and the constructed eddy dataset are publicly available.

Proteomic signatures of Alzheimer's disease and Lewy body dementias: A comparative analysis.

We aimed to identify unique proteomic signatures of Alzheimer's disease (AD), dementia with Lewy bodies (DLB), and Parkinson's disease dementia (PDD). We conducted a comparative proteomic analysis of 33 post mortem brains from AD, DLB, andPDD individuals without dementia focusing on prefrontal, cingulate, and parietal cortices, using weighted gene co-expression network analyses with differential enrichment analysis. Network modules revealed hub proteins common to all dementias. Lewy body dementias differed from AD by reduced levels of the autophagy protein p62 (SQSTM1), whereas DLB was distinguished from both AD and PDD by altered TRIM33 and cysteine/glutamate transporter (SLC7A11) across brain regions. An increase in mitochondrial and synaptic proteins was related to better cognition whereas enrichment in theextracellular matrix, complement system, and autophagy proteins was associated with greater cognitive impairment. Our study offers valuable insights into the network-based biomarker characterization of molecular signatures of AD, DLB, and PDD. Reduced levels of the autophagy protein p62 (SQSTM1) differentiated Lewy body dementias from Alzheimer's disease (AD) across multiple brain regions. Dementia with Lewy bodies (DLB) was distinguished from both AD and Parkinson's disease dementia (PDD) by altered TRIM33 and cysteine/glutamate transporter (SLC7A11) levels across brain regions. Key mitochondrial oxidative phosphorylation proteins (e.g., COX7A2, TOMM40L, NDUFV1), and synaptic proteins (e.g., GABRB3, GABRB2, GLUA3, GLUA4, SNAP47, dynamin1) were more abundant in preserved cognitive states. Extracellular matrix proteins and members of the complement system (decorin, biglycan, C4A, C4B) showed astrong positive correlation with cognitive decline.