Abundant Features Research Articles

Abstract 2034: Gut Microbial Markers Associated With Clinical Features Of Peripheral Artery Disease

Introduction: The human gut microbiome and microbe-derived metabolites are known to be associated with cardiovascular disease. However, their associations with lower extremity peripheral artery disease (PAD) are not well understood. The goal of this study is to identify specific gut microbial markers associated with clinical features of PAD. Methods: Study participants aged ≥ 18 years with documented PAD and non-PAD controls were recruited from a single center and performed home fecal sample collection. Sample aliquots were stored at -80°C until DNA extraction and shotgun metagenome sequencing. Untargeted metabolite quantification was performed using nuclear magnetic resonance. Samples from participants on antibiotics were excluded. Raw paired end sequences were quality filtered using the KneadData pipeline followed by taxonomic annotation using Kraken2 against the UHGG database (v2.0.2) and species abundance estimation using Bracken. A species network was constructed with centered log-ratio transformed abundance data and statistical analyses were performed in R. Results: Among 115 study participants, 80 had PAD and 35 were non-PAD controls. Of those with PAD, 40 (50.0%) had chronic limb threatening ischemia (CLTI) and 39 (48.8%) had symptoms of claudication. There were significant differences in alpha and beta diversity between the claudication and CLTI subgroups but none between the non-PAD and PAD groups overall. There were multiple significantly differentially abundant microbial features in the PAD group compared to non-PAD controls, including species within the genera Blautia and Clostridium . Metabolite profiles were also significantly different between the PAD and non-PAD groups, including higher abundance of the short chain fatty acid butyrate in the non-PAD controls (p=0.02). Network analyses identified three clusters of species with higher relative abundance in the claudication group compared to the CLTI group, and members of these clusters included putative butyrate producers. Conclusions: We identified specific microbial and metabolomic patterns associated with the clinical presentation of PAD and highlight a potential role for microbially-mediated butyrate in disease state.

Deep-Representation-Learning-Based Classification Strategy for Anticancer Peptides

Cancer, with its complexity and numerous origins, continues to provide a huge challenge in medical research. Anticancer peptides are a potential treatment option, but identifying and synthesizing them on a large scale requires accurate prediction algorithms. This study presents an intuitive classification strategy, named ACP-LSE, based on representation learning, specifically, a deep latent-space encoding scheme. ACP-LSE can demonstrate notable advancements in classification outcomes, particularly in scenarios with limited sample sizes and abundant features. ACP-LSE differs from typical black-box approaches by focusing on representation learning. Utilizing an auto-encoder-inspired network, it embeds high-dimensional features, such as the composition of g-spaced amino acid pairs, into a compressed latent space. In contrast to conventional auto-encoders, ACP-LSE ensures that the learned feature set is both small and effective for classification, giving a transparent alternative. The suggested approach is tested on benchmark datasets and demonstrates higher performance compared to the current methods. The results indicate improved Matthew’s correlation coefficient and balanced accuracy, offering insights into crucial aspects for developing new ACPs. The implementation of the proposed ACP-LSE approach is accessible online, providing a valuable and reproducible resource for researchers in the field.

A global survey of host, aquatic, and soil microbiomes reveals shared abundance and genomic features between bacterial and fungal generalists

Environmental change, coupled with alteration in human lifestyles, is profoundly impacting the microbial communities critical to the health of the Earth and its inhabitants. To identify bacteria and fungi that are resistant and susceptible to habitat change, we analyze thousands of genera detected in 1,580 host, soil, and aquatic samples. This large-scale analysis identifies 48 bacterial and 4 fungal genera that are abundant across the three biomes, demonstrating fitness in diverse environmental conditions. Samples containing these generalists have significantly higher alpha diversity. These generalists play a significant role in shaping cross-kingdom community structure, boasting larger genomes with more secondary metabolism and antimicrobial resistance genes. Conversely, 30 bacterial and 19 fungal genera are only found in a single habitat, suggesting a limited ability to adapt to different and changing environments. These findings contribute to our understanding of microbial niche breadth and its consequences for global biodiversity loss.

Variable importance and scale of influence across individual scottish wildcat hybrid habitat models

Understanding the scale dependence of species-habitat relationships is an important area of research in species distribution modeling. There has been little research focused on how habitat selection may depend on individual variation among organisms, geographical location and ecological context of that location. Furthermore, little is known about the extent and drivers of heterogeneity of scale dependence among individuals of a species inhabiting different ecological contexts, and few studies have compared scale dependence and variable importance in a spatially replicated framework. Two of the most important factors for interpreting habitat relationships models include: (1) the relative importance of variables in the model and (2) the spatial scale at which each variable has the largest influence. Based on the existing evidence we hypothesize a priori that landcover variables will generally be the most important predictors, followed by topography, then soil type (which influence both vegetation and prey), Normalized Difference Vegetation Index (NDVI) as an indicator of total vegetation density and perhaps a proxy for prey density, vegetation cover and rabbit abundance. We also expected a priori that there would be consistent patterns of scale dependence across individual wildcat hybrid models related to different variable groups. We expected topographical features to be selected at broad scales, as they influence broad-scale climatic and ecological conditions. We also expected that land cover classes and vegetation cover density to be selected at relatively broad scales given past research showing land cover generally influences habitat selection at relatively broad scales. We expected NDVI and soil type to be selected at finer scales, as their variation influences the distribution of resources and limiting conditions within landscapes. Finally, we expected that rabbit abundance and linear features would affect wildcat hybrid occurrence at the finest scales, given these are resources and conditions that vary over short distances and strongly influence wildcat and wildcat hybrid behavior and habitat use. Our results were consistent with the hypothesis that there may be consistency regarding which variables or variable groups are most important as predictors of wildcat hybrid occurrence in Scotland. Based on previous research we expected that there would be consistent patterns of scale dependence across individual wildcat hybrid models related to different variable groups. Finally, our results identify a clear and consistent trend of increasing frequency of inclusion of variables at increasingly broad scales. This is a linear trend in frequency of variables retained increasing as the scale increased. This suggests a consistent and monotonic pattern of more frequent retention of variables at increasingly broad scales.

Selective Deep Autoencoder for Unsupervised Feature Selection

In light of the advances in big data, high-dimensional datasets are often encountered. Incorporating them into data-driven models can enhance performance; however, this comes at the cost of high computation and the risk of overfitting, particularly due to abundant redundant features. Identifying an informative subset of the features helps in reducing the dimensionality and enhancing model interpretability. In this paper, we propose a novel framework for unsupervised feature selection, called Selective Deep Auto-Encoder (SDAE). It aims to reduce the number of features used in unlabeled datasets without compromising the quality of information obtained. It achieves this by selecting sufficient features - from the original feature set - capable of representing the entire feature space and reconstructing them. Architecturally, it leverages the use of highly nonlinear latent representations in deep Autoencoders and intrinsically learns, in an unsupervised fashion, the relevant and globally representative subset of features through a customized Selective Layer. Extensive experimental results on three high-dimensional public datasets have shown promising feature selection performance by SDAE in comparison to other existing state-of-the-art unsupervised feature selection methods.

Morphological, chemical and growth patterns characterization in shells of Phorcus species along the northeastern coasts of Tunisia

This work is a characterization of the morphometry, the relative growth and the metal contamination degree of the shells of two species belonging to Phorcus genus along the northeastern rocky coasts of Tunisia. The gastropods were sampled from nine locations during the winter 2017. Our findings suggested a spatial morphometric variability probably linked to the extent of the continental shelf, the coasts’ geomorphology, the dynamics of water masses, the predation and the competition effects. Furthermore, the shell investigation allowed the study of the growth pattern and indicated that the development in shell size was faster than the increase in weight. Moreover, the use of Phorcus shells as bioindicators of long-term metal contamination seems to be of growing interest, due to their abundance, high longevity, easy sampling and ecobiological features, both in a scientific and ecosystem management perspective that aims to the establishment of conservation measures targeting marine coastal environments.

Early Triassic (Griesbachian) spheroids in Beibei, Chongqing, Southwestern China: Characteristics, cause and implications for palaeo‐oceanic conditions

Following the latest Permian mass extinction (LPME), abundant unusual sedimentary features and fabrics were widely distributed in Early Triassic carbonate platforms. As a unique type of carbonate grain, spheroids from the Lower Triassic are infrequently reported, yet differ significantly from giant ooids and oncoids. Griesbachian micritic spheroids are well preserved in the Lower Triassic Feixianguan Formation at the Baimiaozi (BMZ) outcrop in Beibei, Chongqing, southwestern China. These spheroids (diameter: 3.9–10.8 mm) were in a marl layer interbedded with massive oolites. They were deposited in the wave troughs of ripple marks comprising micritic clots, sparry calcite, clay minerals, organic matter and pyrite. A microbial spheroid origin is implied based on the coccoidal microbes (coccoid‐like microspherules and bacterial clump‐like microspherules), numerous pyrite framboids (indications suggest the presence of many sulphate‐reducing bacteria and the high iron content in seawater can promote nitrogen‐fixing cyanobacteria proliferation) and microbial‐derived microspherules. The dark‐coloured matrix between the spheroids primarily included microcrystalline calcite, clay minerals, organic matter and metazoan fossils. The spherical and ellipsoidal shapes of the spheroids suggested rapid accretion and lithification. As a unique carbonate depositional mode, Greisbachian spheroids may have been recorded for a palaeo‐ocean with dysoxic and calcium carbonate supersaturation shortly after the LPME.

Multi-Modal Temporal Hypergraph Neural Network for Flotation Condition Recognition.

Efficient flotation beneficiation heavily relies on accurate flotation condition recognition based on monitored froth video. However, the recognition accuracy is hindered by limitations of extracting temporal features from froth videos and establishing correlations between complex multi-modal high-order data. To address the difficulties of inadequate temporal feature extraction, inaccurate online condition detection, and inefficient flotation process operation, this paper proposes a novel flotation condition recognition method named the multi-modal temporal hypergraph neural network (MTHGNN) to extract and fuse multi-modal temporal features. To extract abundant dynamic texture features from froth images, the MTHGNN employs an enhanced version of the local binary pattern algorithm from three orthogonal planes (LBP-TOP) and incorporates additional features from the three-dimensional space as supplements. Furthermore, a novel multi-view temporal feature aggregation network (MVResNet) is introduced to extract temporal aggregation features from the froth image sequence. By constructing a temporal multi-modal hypergraph neural network, we encode complex high-order temporal features, establish robust associations between data structures, and flexibly model the features of froth image sequence, thus enabling accurate flotation condition identification through the fusion of multi-modal temporal features. The experimental results validate the effectiveness of the proposed method for flotation condition recognition, providing a foundation for optimizing flotation operations.

Studi kelimpahan dan morfometrik pada kerang darah (Anadara granosa) di Perairan Motadikin Kabupaten Malaka

Anadara granosa is one of the fisheries commodities that is abundant in various regions, especially tropical areas. The high use of blood cockles by the community in Motadikin waters is considered to result in a decline in the population of blood cockles in nature, thereby increasing population growth. This research aims to examine the abundance and morphometric features of blood cockles in Motadikin waters, Malaka Regency. This research was carried out for 1 month in Motadikin waters, Malacca Regency using the transect method. The number of transects used was 3, the distance between transects was 25 m and each transect used 6 quadrants with a size of 1x1 m and the distance between each quadrant was 5 m. The results of the research showed that the abundance of blood clams in Motadikin Waters was an average of 5-7 individuals/m2 and the average The average abundance found on substrates covered with mangroves (I) is higher than on sandy substrates alone (II). Meanwhile, morphometric measurements of blood clams in Motadikin waters are dominated by blood clams with a length range of 3.2 cm - 3.9 cm, with a shell height of 1.4 cm - 2.7 cm, an umbo height of 1.1 cm - 1.2 cm, shell thickness 2.4 cm - 2.6 cm and ligament length 2.4 cm - 2.7 cm.
 
 Keywords: Anadara granosa, abundance, morphometrics, motadikin waters.

2D Conjugated Metal-Organic Frameworks: Defined Synthesis and Tailor-Made Functions.

Conspectus2D conjugated metal-organic frameworks (2D -MOFs) have emerged as a class of graphene-like materials with fully π-conjugated aromatic structures. Their unique structural characteristics provide abundant physiochemical features, including regular nanochannels, high electrical conductivity, and customizable band gaps. Recent intensive research has significantly advanced this field, yet the exploration of 2D c-MOFs with enhanced features is limited by the availability of organic linkages and topologies. Designing novel ligands is essential for the construction of new 2D c-MOFs with high crystallinity, excellent conductivity, and tailor-made functions.In this Account, we summarize our recent contributions in fine-tuning the topology of 2D c-MOFs through precise ligand design, thereby giving them fantastic structures and tailor-made functions. First, we propose the concept of replacing planar ligands by nonplanar ligands on conductive MOF skeletons. The incorporation of nonplanar ligands improves the solubility of large π-conjugated organic molecules without interfering with the interlayer π-stacking. Our investigation discovered that conjugate polycyclic aromatics-based ligands can be synthesized through in situ Scholl reactions by means of oxidative cyclodehydrogenation of a nonplanar precursor ligand during the solvothermal synthesis process. Hence, fully conjugated 2D c-MOFs can be directly synthesized from nonplanar organic ligands, simplifying and diversifying the preparation of 2D c-MOFs. Accordingly, the design flexibility of the ligands expands the topological structures and pore types. By controlling the synthesis conditions, we can successfully induce either a rhombus or a kagome topology from a nonplanar D2 symmetric ligand. Moreover, by employing a ligand engineering strategy, we incrementally increase the number of coordination functional groups on a twisted hexabenzocoronene core, resulting in the formation of three distinct symmetric hydroxyl ligands. These ligands elicit diverse target topologies and pore sizes, resulting in variances in the coordination node density on the skeletons. This, in turn, leads to differences in electron transfer abilities, ultimately causing variations in the electrical conductivity and mobility. In addition, we employ a straightforward coupling method to incorporate redox components, such as salphen and pyrazine, into nonplanar ligands, facilitating the synthesis of 2D c-MOFs with highly active centers. This strategy confers upon the resulting frameworks substantial capacity for catalysis and energy storage, offering a good platform for elucidating the structure-property relationships at the molecular level. Moreover, the well-defined synthesis of 2D c-MOFs imparts them with specific properties, particularly in the fields of electrical, electrochemical, and spintronic applications. At the end, the primary challenges facing 2D c-MOFs in achieving tailor functions and their practical applications are proposed. This account is expected to evoke new inspirations and innovative research in the field of 2D c-MOFs, especially in emerging interdisciplinary research areas.

S3-Net: A Self-Supervised Dual-Stream Network for Radiology Report Generation.

Intelligent medicine is eager to automatically generate radiology reports to ease the tedious work of radiologists. Previous researches mainly focused on the text generation with encoder-decoder structure, while CNN networks for visual features ignored the long-range dependencies correlated with textual information. Besides, few studies exploit cross-modal mappings to promote radiology report generation. To alleviate the above problems, we propose a novel end-to-end radiology report generation model dubbed Self-Supervised dual-Stream Network (S3-Net). Specifically, a Dual-Stream Visual Feature Extractor (DSVFE) composed of ResNet and SwinTransformer is proposed to capture more abundant and effective visual features, where the former focuses on local response and the latter explores long-range dependencies. Then, we introduced the Fusion Alignment Module (FAM) to fuse the dual-stream visual features and facilitate alignment between visual features and text features. Furthermore, the Self-Supervised Learning with Mask(SSLM) is introduced to further enhance the visual feature representation ability. Experimental results on two mainstream radiology reporting datasets (IU X-ray and MIMIC-CXR) show that our proposed approach outperforms previous models in terms of language generation metrics.

Predicting the Rapid Progression of Mild Cognitive Impairment by Intestinal Flora and Blood Indicators through Machine Learning Method

Introduction: The aim of the work was to establish a prediction model of mild cognitive impairment (MCI) progression based on intestinal flora by machine learning method. Method: A total of 1,013 patients were recruited, in which 87 patients with MCI finished a two-year follow-up. To establish a prediction model, 61 patients were randomly divided into a training set and 26 patients were divided into a testing set. A total of 121 features including demographic characteristics, hematological indicators, and intestinal flora abundance were analyzed. Results: Of the 87 patients who finished a two-year follow-up, 44 presented rapid progression. Model 1 was established based on 121 features with the accuracy 85%, sensitivity 85%, and specificity 83%. Model 2 was based on the first fifteen features of model 1 (triglyceride, uric acid, alanine transaminase, F-Clostridiaceae, G-Megamonas, S-Megamonas, G-Shigella, G-Shigella, S-Shigella, average hemoglobin concentration, G-Alistipes, S-Collinsella, median cell count, average hemoglobin volume, low-density lipoprotein), with the accuracy 97%, sensitivity 92%, and specificity 100%. Model 3 was based on the first ten features of model 1, with the accuracy 97%, sensitivity 86%, and specificity 100%. Other models based on the demographic characteristics, hematological indicators, or intestinal flora abundance features presented lower sensitivity and specificity. Conclusion: The 15 features (including intestinal flora abundance) could establish an effective model for predicting rapid MCI progression.

Boundary-Rich Carbon-Based Electrocatalysts with Manganese(II)-Coordinated Active Environment for Selective Synthesis of Hydrogen Peroxide.

Coordinated manganese (Mn) electrocatalysts owing to their electronic structure flexibility, non-toxic and earth abundant features are promising for electrocatalytic reactions. However, achieving selective hydrogen peroxide (H2 O2 ) production through two electron oxygen reduction (2e-ORR) is a challenge on Mn-centered catalysts. Targeting this goal, we report on the creation of a secondary Mn(II)-coordinated active environment with reactant enrichment effect on boundary-rich porous carbon-based electrocatalysts, which facilitates the selective and rapid synthesis of H2 O2 through 2e-ORR. The catalysts exhibit nearly 100 % Faradaic efficiency and H2 O2 productivity up to 15.1 mol gcat -1 h-1 at 0.1 V versus reversible hydrogen electrode, representing the record high activity for Mn-based electrocatalyst in H2 O2 electrosynthesis. Mechanistic studies reveal that the epoxide and hydroxyl groups surrounding Mn(II) centers improve spin state by modifying electronic properties and charge transfer, thus tailoring the adsorption strength of *OOH intermediate. Multiscale simulations reveal that the high-curvature boundaries facilitate oxygen (O2 ) adsorption and result in local O2 enrichment due to the enhanced interaction between carbon surface and O2 . These merits together ensure the efficient formation of H2 O2 with high local concentration, which can directly boost the tandem reaction of hydrolysis of benzonitrile to benzamide with nearly 100 % conversion rate and exclusive benzamide selectivity.

Boundary‐Rich Carbon‐Based Electrocatalysts with Manganese(II)‐Coordinated Active Environment for Selective Synthesis of Hydrogen Peroxide

AbstractCoordinated manganese (Mn) electrocatalysts owing to their electronic structure flexibility, non‐toxic and earth abundant features are promising for electrocatalytic reactions. However, achieving selective hydrogen peroxide (H2O2) production through two electron oxygen reduction (2e‐ORR) is a challenge on Mn‐centered catalysts. Targeting this goal, we report on the creation of a secondary Mn(II)‐coordinated active environment with reactant enrichment effect on boundary‐rich porous carbon‐based electrocatalysts, which facilitates the selective and rapid synthesis of H2O2 through 2e‐ORR. The catalysts exhibit nearly 100 % Faradaic efficiency and H2O2 productivity up to 15.1 mol gcat−1 h−1 at 0.1 V versus reversible hydrogen electrode, representing the record high activity for Mn‐based electrocatalyst in H2O2 electrosynthesis. Mechanistic studies reveal that the epoxide and hydroxyl groups surrounding Mn(II) centers improve spin state by modifying electronic properties and charge transfer, thus tailoring the adsorption strength of *OOH intermediate. Multiscale simulations reveal that the high‐curvature boundaries facilitate oxygen (O2) adsorption and result in local O2 enrichment due to the enhanced interaction between carbon surface and O2. These merits together ensure the efficient formation of H2O2 with high local concentration, which can directly boost the tandem reaction of hydrolysis of benzonitrile to benzamide with nearly 100 % conversion rate and exclusive benzamide selectivity.

Dual-Branch Multi-Scale Relation Networks with Tutorial Learning for Few-Shot Learning

Few-shot learning refers to training a model with a few labeled data to effectively recognize unseen categories. Recently, numerous approaches have been suggested to improve the extraction of abundant feature information at hierarchical layers or multiple scales for similarity metrics, especially methods based on learnable relation networks, which have demonstrated promising results. However, the roles played by image features in relationship measurement vary at different layers, and effectively integrating features from different layers and multiple scales can improve the measurement capacity of the model. In light of this, we propose a novel method called dual-branch multi-scale relation networks with tutoring learning (DbMRNT) for few-shot learning. Specifically, we first generate deep multiple features using a multi-scale feature generator in Branch 1 while extracting features at hierarchical layers in Branch 2. Then, learnable relation networks are employed in both branches to measure the pairwise similarity of features at each scale or layer. Furthermore, to leverage the dominant role of deep features in the final classification, we introduce a tutorial learning module that enables Branch 1 to tutor the learning process of Branch 2. Ultimately, the relation scores of all scales and layers are integrated to obtain the classification results. Extensive experiments on popular few-shot learning datasets prove that our method outperforms other similar methods.

The distribution characteristics of aerosol bacteria in different types of sheepfolds.

With the development of modern sheep raising technology, the increasing density of animals in sheep house leads to the accumulation of microbial aerosols in sheep house. It is an important prerequisite to grasp the characteristics of bacteria in aerosols in sheep house to solve the problems of air pollution and disease prevention and control in sheep house. In this study, the microorganisms present in the air of sheep houses were investigated to gain insights into the structure of bacterial communities and the prevalence of pathogenic bacteria. Samples from six sheep pens in each of three sheep farms, totaling 18, were collected in August 2022 from Ningxia province, China. A high-volume air sampler was utilized for aerosol collection within the sheep housing followed by DNA extraction for 16S rRNA sequencing. Employing high-throughput 16S rRNA sequencing technology, we conducted an in-depth analysis of microbial populations in various sheep pen air samples, enabling us to assess the community composition and diversity. The results revealed a total of 11,207 operational taxonomic units (OTUs) within the bacterial population across the air samples, encompassing 152 phyla, 298 classes, 517 orders, 853 families, 910 genera, and 482 species. Alpha diversity and beta diversity analysis indicated that differences in species diversity, evenness and coverage between different samples. At the bacterial phylum level, the dominant bacterial groups are Firmicutes, Proteobacteria, and Actinobacteria, among which Firmicutes (97.90-98.43%) is the highest. At the bacterial genus level, bacillus, Bacteroides, Fusobacterium, etc. had higher abundance, with Bacillus (85.47-89.87%) being the highest. Through an in-depth analysis of microbial diversity and a meticulous examination of pathogenic bacteria with high abundance in diverse sheep house air samples, the study provided valuable insights into the microbial diversity, abundance, and distinctive features of prevalent pathogenic bacteria in sheep house air. These findings serve as a foundation for guiding effective disease prevention and control strategies within sheep farming environments.

Marine oil spill detection using improved polarimetric feature based on polarization SAR image

ABSTRACT Monitoring marine oil pollution holds both practical and scientific significance. Synthetic Aperture Radar (SAR) is an active microwave remote sensing technique capable of all-weather and all-day with fine spatial resolution. However, under low wind conditions, rain cells and young ice are examples of look-alikes affect the accuracy of oil spill detection. Polarimetric SAR assumes a crucial role in this context, as it can extract abundant polarimetric features by polarimetric target decomposition. Drawing inspiration from this advancement, an improved polarimetric feature named relative feature based on Cloude-Pottier target decomposition was proposed. The Jeffries-Matusita distance indicates the substantial potential of the relative feature in detecting oil spills. The improved polarimetric feature within U-Net, FCN-8s, and DeepLabv3+ResNet-18 for oil spill detection using polarization SAR images. Experiment results demonstrated that the relative feature has superior performance compared to other polarimetric features and obtained the highest accuracy and dice within U-Net compared to the other two models. These findings introduce promising concepts for achieving rapid and precise detection of oil spills in future applications.

A dual-ways feature fusion mechanism enhancing active learning based on TextCNN

Active Learning (AL) is a technique being widely employed to minimize the time and labor costs in the task of annotating data. By querying and extracting the specific instances to train the model, the relevant task’s performance is improved maximally within limited iterations. However, rare work was conducted to fully fuse features from different hierarchies to enhance the effectiveness of active learning. Inspired by the thought of information compensation in many famous deep learning models (such as ResNet, etc.), this work proposes a novel TextCNN-based Two ways Active Learning model (TCTWAL) to extract task-relevant texts. TextCNN takes the advantage of little hyper-parameter tuning and static vectors and achieves excellent results on various natural language processing (NLP) tasks, which are also beneficial to human-computer interaction (HCI) and the AL relevant tasks. In the process of the proposed AL model, the candidate texts are measured from both global and local features by the proposed AL framework TCTWAL depending on the modified TextCNN. Besides, the query strategy is strongly enhanced by maximum normalized log-probability (MNLP), which is sensitive to detecting the longer sentences. Additionally, the selected instances are characterized by general global information and abundant local features simultaneously. To validate the effectiveness of the proposed model, extensive experiments are conducted on three widely used text corpus, and the results are compared with with eight manual designed instance query strategies. The results show that our method outperforms the planned baselines in terms of accuracy, macro precision, macro recall, and macro F1 score. Especially, to the classification results on AG’s News corpus, the improvements of the four indicators after 39 iterations are 40.50%, 45.25%, 48.91%, and 45.25%, respectively.

Mn2+ recycling in hypersaline wastewater: unnoticed intracellular biomineralization and pre-cultivation of immobilized bacteria.

Microbially induced manganese carbonate precipitation has been utilized for the treatment of wastewater containing manganese. In this study, Virgibacillus dokdonensis was used to remove manganese ions from an environment containing 5% NaCl. The results showed a significant decrease in carbonic anhydrase activity and concentrations of carbonate and bicarbonate ions with increasing manganese ion concentrations. However, the levels of humic acid analogues, polysaccharides, proteins, and DNA in EPS were significantly elevated compared to those in a manganese-free environment. The rhodochrosite exhibited a preferred growth orientation, abundant morphological features, organic elements including nitrogen, phosphorus, and sulfur, diverse protein secondary structures, as well as stable carbon isotopes displaying a stronger negative bias. The presence of manganese ions was found to enhance the levels of chemical bonds O-C=O and N-C=O in rhodochrosite. Additionally, manganese in rhodochrosite exhibited both + 2 and + 3 valence states. Rhodochrosite forms not only on the cell surface but also intracellularly. After being treated with free bacteria for 20days, the removal efficiency of manganese ions ranged from 88.4 to 93.2%, and reached a remarkable 100% on the 10th day when using bacteria immobilized on activated carbon fiber that had been pre-cultured for three days. The removal efficiency of manganese ions was significantly enhanced under the action of pre-cultured immobilized bacteria compared to non-pre-cultured immobilized bacteria. This study contributes to a comprehensive understanding of the mineralization mechanism of rhodochrosite, thereby providing an economically and environmentally sustainable biological approach for treating wastewater containing manganese.

C2BNet: A Deep Learning Architecture with Coupled Composite Backbone for Parasitic EGG Detection in Microscopic Images.

Internet of Medical Things (IoMT) enabled by artificial intelligence (AI) technologies can facilitate automatic diagnosis and management of chronic diseases (e.g., intestinal parasitic infection) based on two-dimensional (2D) microscopic images. To improve the model performance of object detection challenged by microscopic image characteristics (e.g., focus failure, motion blur, and whether zoomed or not), we propose Coupled Composite Backbone Network (C2BNet) to execute the parasitic egg detection task using 2D microscopic images. In particular, the C2BNet backbone adopts a two-path structure-based backbone and leverages model heterogeneity to learn object features from different perspectives. A novel feature composition style is proposed to flow the feature within the coupled composite backbone, and ensure mutual enhancement of feature representation ability among the different paths of the backbone. To further improve the accuracy of the detection results, we propose Multiscale Weighted Box Fusion (WBF) to fuse the location and confidence scores of all bounding boxes predicted from the multiscale feature maps, and iteratively refine the box coordinates to form the final prediction. Experimental results on Chula-ParasiteEgg-11 dataset demonstrate that the C2BNet not only performs satisfactorily compared with state-of-the-art methods, but also can focus more on learning detailed morphology features and abundant semantic features, resulting in more precise detection for parasitic eggs located in the 2D microscopic image.