CE-GAN: Community Evolutionary Generative Adversarial Network for Alzheimer's Disease Risk Prediction.

In the studies of neurodegenerative diseases such as Alzheimer's Disease (AD), researchers often focus on the associations among multi-omics pathogeny based on imaging genetics data. However, current studies overlook the communities in brain networks, leading to inaccurate models of disease development. This paper explores the developmental patterns of AD from the perspective of community evolution. We first establish a mathematical model to describe functional degeneration in the brain as the community evolution driven by entropy information propagation. Next, we propose an interpretable Community Evolutionary Generative Adversarial Network (CE-GAN) to predict disease risk. In the generator of CE-GAN, community evolutionary convolutions are designed to capture the evolutionary patterns of AD. The experiments are conducted using functional magnetic resonance imaging (fMRI) data and single nucleotide polymorphism (SNP) data. CE-GAN achieves 91.67% accuracy and 91.83% area under curve (AUC) in AD risk prediction tasks, surpassing advanced methods on the same dataset. In addition, we validated the effectiveness of CE-GAN for pathogeny extraction. The source code of this work is available at https://github.com/fmri123456/CE-GAN.

Identifying the Local Influencing Factors of Arsenic Concentration in Suburban Soil: A Multiscale Geographically Weighted Regression Approach.

Exploring the local influencing factors and sources of soil arsenic (As) is crucial for reducing As pollution, protecting soil ecology, and ensuring human health. Based on geographically weighted regression (GWR), multiscale GWR (MGWR) considers the different influence ranges of explanatory variables and thus adopts an adaptative bandwidth. It is an effective model in many fields but has not been used in exploring local influencing factors and sources of As. Therefore, using 200 samples collected from the northeastern black soil zone of China, this study examined the effectiveness of MGWR, revealed the spatial non-stationary relationship between As and environmental variables, and determined the local impact factors and pollution sources of As. The results showed that 49% of the samples had arsenic content exceeding the background value, and these samples were mainly distributed in the central and southern parts of the region. MGWR outperformed GWR with the adaptative bandwidth, with a lower Moran's I of residuals and a higher R2 (0.559). The MGWR model revealed the spatially heterogeneous relationship between As and explanatory variables. Specifically, the road density and total nitrogen, clay, and silt contents were the primary or secondary influencing factors at most points. The distance from an industrial enterprise was the secondary influencing factor at only a few points. The main pollution sources of As were thus inferred as traffic and fertilizer, and industrial emissions were also included in the southern region. These findings highlight the importance of considering adaptative bandwidths for independent variables and demonstrate the effectiveness of MGWR in exploring local sources of soil pollutants.

Role of CD44 in Chemotherapy Treatment Outcome: A Scoping Review of Clinical Studies.

Cluster of differentiation 44 (CD44), a cell surface adhesion molecule overexpressed in cancer stem cells, has been implicated in chemoresistance. This scoping review, following PRISMA-ScR guidelines, systematically identified and evaluated clinical studies on the impact of CD44 expression on chemotherapy treatment outcomes across various cancer types. The search encompassed PubMed (1985-2023) and SCOPUS (1936-2023) databases, yielding a total of 12,659 articles, of which 40 met the inclusion criteria and were included in the qualitative synthesis using a predefined data extraction table. Data collected included the cancer type, sample size, interventions, control, treatment outcome, study type, expression of CD44 variants and isoforms, and effect of CD44 on chemotherapy outcome. Most of the studies demonstrated an association between increased CD44 expression and negative chemotherapeutic outcomes such as shorter overall survival, increased tumor recurrence, and resistance to chemotherapy, indicating a potential role of CD44 upregulation in chemoresistance in cancer patients. However, a subset of studies also reported non-significant relationships or conflicting results. In summary, this scoping review highlighted the breadth of the available literature investigating the clinical association between CD44 and chemotherapeutic outcomes. Further research is required to elucidate this relationship to aid clinicians in managing CD44-positive cancer patients.

Hierarchical functional differences between gyri and sulci at different scales.

Gyri and sulci are 2 fundamental cortical folding patterns of the human brain. Recent studies have suggested that gyri and sulci may play different functional roles given their structural and functional heterogeneity. However, our understanding of the functional differences between gyri and sulci remains limited due to several factors. Firstly, previous studies have typically focused on either the spatial or temporal domain, neglecting the inherently spatiotemporal nature of brain functions. Secondly, analyses have often been restricted to either local or global scales, leaving the question of hierarchical functional differences unresolved. Lastly, there has been a lack of appropriate analytical tools for interpreting the hierarchical spatiotemporal features that could provide insights into these differences. To overcome these limitations, in this paper, we proposed a novel hierarchical interpretable autoencoder (HIAE) to explore the hierarchical functional difference between gyri and sulci. Central to our approach is its capability to extract hierarchical features via a deep convolutional autoencoder and then to map these features into an embedding vector using a carefully designed feature interpreter. This process transforms the features into interpretable spatiotemporal patterns, which are pivotal in investigating the functional disparities between gyri and sulci. We evaluate the proposed framework on Human Connectome Project task functional magnetic resonance imaging dataset. The experiments demonstrate that the HIAE model can effectively extract and interpret hierarchical spatiotemporal features that are neuroscientifically meaningful. The analyses based on the interpreted features suggest that gyri are more globally activated, whereas sulci are more locally activated, demonstrating a distinct transition in activation patterns as the scale shifts from local to global. Overall, our study provides novel insights into the brain's anatomy-function relationship.

Technical note: Generalizable and promptable artificial intelligence model to augment clinical delineation in radiation oncology.

BACKGROUND: Efficient and accurate delineation of organs at risk (OARs) is a critical procedure for treatment planning and dose evaluation. Deep learning-based auto-segmentation of OARs has shown promising results and is increasingly being used in radiation therapy. However, existing deep learning-based auto-segmentation approaches face two challenges in clinical practice: generalizability and human-AI interaction. A generalizable and promptable auto-segmentation model, which segments OARs of multiple disease sites simultaneously and supports on-the-fly human-AI interaction, can significantly enhance the efficiency of radiation therapy treatment planning. PURPOSE: Meta's segment anything model (SAM) was proposed as a generalizable and promptable model for next-generation natural image segmentation. We further evaluated the performance of SAM in radiotherapy segmentation. METHODS: Computed tomography (CT) images of clinical cases from four disease sites at our institute were collected: prostate, lung, gastrointestinal, and head & neck. For each case, we selected the OARs important in radiotherapy treatment planning. We then compared both the Dice coefficients and Jaccard indices derived from three distinct methods: manual delineation (ground truth), automatic segmentation using SAM's 'segment anything' mode, and automatic segmentation using SAM's 'box prompt' mode that implements manual interaction via live prompts during segmentation. RESULTS: Our results indicate that SAM's segment anything mode can achieve clinically acceptable segmentation results in most OARs with Dice scores higher than 0.7. SAM's box prompt mode further improves Dice scores by 0.1∼0.5. Similar results were observed for Jaccard indices. The results show that SAM performs better for prostate and lung, but worse for gastrointestinal and head & neck. When considering the size of organs and the distinctiveness of their boundaries, SAM shows better performance for large organs with distinct boundaries, such as lung and liver, and worse for smaller organs with less distinct boundaries, like parotid and cochlea. CONCLUSIONS: Our results demonstrate SAM's robust generalizability with consistent accuracy in automatic segmentation for radiotherapy. Furthermore, the advanced box-prompt method enables the users to augment auto-segmentation interactively and dynamically, leading to patient-specific auto-segmentation in radiation therapy. SAM's generalizability across different disease sites and different modalities makes it feasible to develop a generic auto-segmentation model in radiotherapy.

Impacts of the aerosol mixing state and new particle formation on CCN in summer at the summit of Mount Tai (1534m) in Central East China.

In this study, the aerosol size distributions, cloud condensation nuclei (CCN) number concentration (NCCN), single-particle chemical composition and meteorological data were collected from May 12 to June 8, 2017, at the summit of Mt. Tai. The effects of new particle formation (NPF) events and aerosol chemical components on CCN at Mt. Tai were analyzed in detail. The results showed that, NPF events significantly enhanced the CCN population, and the enhancement effect increased with increasing supersaturation (SS) value at Mt.Tai. NCCN at SS ranging from 0.1 to 0.9 % on NPF days was 10.9 %, 36.5 %, 44.6 %, 53.5 % and 51.5 % higher than that on non-NPF days from 10:00-13:00 as NPF events progressed. The effect of chemical components on CCN activation under the influence of NPF events was greater than that in the absence of NPF events. The correlation coefficients of EC-Nitrate particles (EC-Sulfate particles) and CCN at all SS levels on NPF days were 1.31-1.59 times (1.17-1.35 times) higher than those on non-NPF days. Nitrate particles promoted CCN activation but sulfate particles inhibited activation at Mt. Tai. There are differences or even opposite effects of the same group of particles on CCN activation under the influence of NPF events in different air masses. EC-Sulfate particles inhibited CCN activation at all SS levels for type I but weakly promoted activation at lower SS ranging from 0.1 to 0.3 % and weakly inhibited it at higher 0.9 % SS for type II. OCEC particles significantly inhibited CCN activation for type II, and this effect decreased with increasing SS. OCEC particles only weakly inhibited activation at SS ranging from 0.5 to 0.7 % for type I. OCEC particles only weakly inhibited this process at 0.1 % SS, while they very weakly promoted activation for SS > 0.1 %. This reveals that the CCN activity is not only related to the chemical composition of the particles, but the mixing state also has an important effect on the CCN activity.

A Unified and Biologically Plausible Relational Graph Representation of Vision Transformers.

Vision transformer (ViT) and its variants have achieved remarkable success in various tasks. The key characteristic of these ViT models is to adopt different aggregation strategies of spatial patch information within the artificial neural networks (ANNs). However, there is still a key lack of unified representation of different ViT architectures for systematic understanding and assessment of model representation performance. Moreover, how those well-performing ViT ANNs are similar to real biological neural networks (BNNs) is largely unexplored. To answer these fundamental questions, we, for the first time, propose a unified and biologically plausible relational graph representation of ViT models. Specifically, the proposed relational graph representation consists of two key subgraphs: an aggregation graph and an affine graph. The former considers ViT tokens as nodes and describes their spatial interaction, while the latter regards network channels as nodes and reflects the information communication between channels. Using this unified relational graph representation, we found that: 1) model performance was closely related to graph measures; 2) the proposed relational graph representation of ViT has high similarity with real BNNs; and 3) there was a further improvement in model performance when training with a superior model to constrain the aggregation graph.

An urban metabolism framework purpose to assess the dynamics of linear alkylbenzenes (LABs) discharge flows and driving forces at the city level in Guangdong province, China.

Linear alkylbenzenes (LABs) are a class of molecular markers derived from anthropogenic activities. A comprehensive understanding of the mechanism that determines their entry into anthroposphere, in terms of magnitude and pathway, is the prerequisite to establish effective mitigation measures. This study develops a methodology framework to analyze the source-sink interactions and driving factors of the direct and indirect LAB discharges from production and living activities in Guangdong Province, China from 2004 to 2017. Results indicated that the total LAB discharges of Guangdong into the environment were averaged at 2.9 kt yr-1, of which 61.9% originated from the Pearl River Delta (PRD) urban agglomeration. An average proportion of 76.0% was discharged into water bodies with the remaining released into land bodied. From 2014 to 2017, the LAB discharges increased by seven times, resulting from the steady increase of urban residential sources, while contribution from industrial sources continuously declined during the studied period. Meanwhile, the discharging hotspots expanded from Guangzhou city to other super-cities around it, including Shenzhen and Dongguan. The other cities exhibited a decreasing trend in discharges as a function of distance from these hotspot cities. The multisectoral sources of LABs differed considerably among cities, and the source contribution of each city changed significantly with progressive urbanization. The factor decomposition analysis indicated that LAB discharges in PRD cities primarily contributed by the pollutant concentration and reflected the treatment structure, while peripheral cities of the PRD mainly contributed by the per capita consumption and pollutant discharge per unit of GDP. Overall, our results provide a scientific database and supports for the regional co-remediation of anthropogenic pollution.

Dual Heterojunctions and Nanobowl Morphology Engineered BiVO4 Photoanodes for Enhanced Solar Water Splitting.

Photoelectrochemical (PEC) water splitting represents an attractive strategy to realize the conversion from solar energy to hydrogen energy, but severe charge recombination in photoanodes significantly limits the conversion efficiency. Herein, a unique BiVO4 (BVO) nanobowl (NB) heterojunction photoanode, which consists of [001]-oriented BiOCl underlayer and BVO nanobowls containing embedded BiOCl nanocrystals, is fabricated by nanosphere lithography followed by in situ transformation. Experimental characterizations and theoretical simulation prove that nanobowl morphology can effectively enhance light absorption while reducing carrier diffusion path. Density functional theory (DFT) calculations show the tendency of electron transfer from BVO to BiOCl. The [001]-oriented BiOCl underlayer forms a compact type II heterojunction with the BVO, favoring electron transfer from BVO through BiOCl to the substrate. Furthermore, the embedded BiOCl nanoparticles form a bulk heterojunction to facilitate bulk electron transfer. Consequently, the dual heterojunctions engineered BVO/BiOCl NB photoanode exhibits attractive PEC performance toward water oxidation with an excellent bulk charge separation efficiency of 95.5%, and a remarkable photocurrent density of 3.38 mA cm-2 at 1.23 V versus reversible hydrogen electrode, a fourfold enhancement compared to the flat BVO counterpart. This work highlights the great potential of integrating dual heterojunctions engineering and morphology engineering in fabricating high-performance photoelectrodes toward efficient solar conversion.

SOX10 deficiency-mediated LAMB3 upregulation determines the invasiveness of MAPKi-resistant melanoma.

Melanoma that develops adaptive resistance to MAPK inhibitors (MAPKi) through transcriptional reprograming-mediated phenotype switching is associated with enhanced metastatic potential, yet the underlying mechanism of this improved invasiveness has not been fully elucidated. In this study, we show that MAPKi-resistant melanoma cells are more motile and invasive than the parental cells. We further show that LAMB3, a ß subunit of the extracellular matrix protein laminin-332 is upregulated in MAPKi-resistant melanoma cells and that the LAMB3-Integrin α3/α6 signaling mediates the motile and invasive phenotype of resistant cells. In addition, we demonstrate that SOX10 deficiency in MAPKi-resistant melanoma cells drives LAMB3 upregulation through TGF-ß signaling. Transcriptome profiling and functional studies further reveal a FAK/MMPs axis mediates the pro-invasiveness effect of LAMB3. Using a mouse lung metastasis model, we demonstrate LAMB3 depletion inhibits the metastatic potential of MAPKi-resistant cells in vivo. In summary, this study identifies a SOX10low/TGF-ß/LAMB3/FAK/MMPs signaling pathway that determines the migration and invasion properties of MAPKi-resistant melanoma cells and provide rationales for co-targeting LAMB3 to curb the metastasis of melanoma cells in targeted therapy.

Differentiating ChatGPT-Generated and Human-Written Medical Texts: Quantitative Study.

BACKGROUND: Large language models, such as ChatGPT, are capable of generating grammatically perfect and human-like text content, and a large number of ChatGPT-generated texts have appeared on the internet. However, medical texts, such as clinical notes and diagnoses, require rigorous validation, and erroneous medical content generated by ChatGPT could potentially lead to disinformation that poses significant harm to health care and the general public. OBJECTIVE: This study is among the first on responsible artificial intelligence-generated content in medicine. We focus on analyzing the differences between medical texts written by human experts and those generated by ChatGPT and designing machine learning workflows to effectively detect and differentiate medical texts generated by ChatGPT. METHODS: We first constructed a suite of data sets containing medical texts written by human experts and generated by ChatGPT. We analyzed the linguistic features of these 2 types of content and uncovered differences in vocabulary, parts-of-speech, dependency, sentiment, perplexity, and other aspects. Finally, we designed and implemented machine learning methods to detect medical text generated by ChatGPT. The data and code used in this paper are published on GitHub. RESULTS: Medical texts written by humans were more concrete, more diverse, and typically contained more useful information, while medical texts generated by ChatGPT paid more attention to fluency and logic and usually expressed general terminologies rather than effective information specific to the context of the problem. A bidirectional encoder representations from transformers-based model effectively detected medical texts generated by ChatGPT, and the F1 score exceeded 95%. CONCLUSIONS: Although text generated by ChatGPT is grammatically perfect and human-like, the linguistic characteristics of generated medical texts were different from those written by human experts. Medical text generated by ChatGPT could be effectively detected by the proposed machine learning algorithms. This study provides a pathway toward trustworthy and accountable use of large language models in medicine.

Enhancement effect of carvacrol on yeast inactivation by mild pressure carbon dioxide.

Saccharomyces cerevisiae is one of the common spoilage microorganisms in fruit juices. This paper investigated the influences of carvacrol on S. cerevisiae inactivation by mild pressure carbon dioxide (MPCO2). The results demonstrated that carvacrol synergistically enhanced the antifungal activity against S. cerevisiae of MPCO2. With the increase of carvacrol concentration (20-160 µg/mL), CO2 pressure (1.5-3.5 MPa), process temperature (20-40 °C), and treatment time (15-60 min), the inactivation effect of carvacrol combined with MPCO2 on S. cerevisiae was gradually increased and significantly stronger than either single treatment. In the presence of carvacrol, MPCO2 severely disordered the plasma membrane of S. cerevisiae, including the increase of membrane permeability, and the loss of membrane potential and integrity. MPCO2 and carvacrol in combination also aggravated the mitochondrial depolarization of S. cerevisiae and reduced intracellular ATP and protein content. This study suggests the potential of carvacrol and pressurized CO2 as an alternative technology for food pasteurization.

A Self-Powered UV Photodetector With Ultrahigh Responsivity Based on 2D Perovskite Ferroelectric Films With Mixed Spacer Cations.

Self-powered photodetectors (PDs) have the advantages of no external power requirement, wireless operation, and long life. Spontaneous ferroelectric polarizations can significantly increase built-in electric field intensity, showing great potential in self-powered photodetection. Moreover, ferroelectrics possess pyroelectric and piezoelectric properties, beneficial for enhancing self-powered PDs. 2D metal halide perovskites (MHPs), which have ferroelectric properties, are suitable for fabricating high-performance self-powered PDs. However, the research on 2D metal halide perovskites ferroelectrics focuses on growing bulk crystals. Herein, 2D ferroelectric perovskite films with mixed spacer cations for self-powered PDs are demonstrated by mixing Ruddlesden-Popper (RP)-type and Dion-Jacobson (DJ)-type perovskite. The (BDA0.7 (BA2 )0.3 )(EA)2 Pb3 Br10 film possesses, overall, the best film qualities with the best crystalline quality, lowest trap density, good phase purity, and obvious ferroelectricity. Based on the ferro-pyro-phototronic effect, the PD at 360 nm exhibits excellent photoelectric properties, with an ultrahigh peak responsivity greater than 93 A W-1 and a detectivity of 2.5 × 1015 Jones, together with excellent reproducibility and stability. The maximum responsivities can be modulated by piezo-phototronic effect with an effective enhancement ratio of 480%. This work will open up a new route of designing MHP ferroelectric films for high-performance PDs and offers the opportunity to utilize it for various optoelectronics applications.

Differences of Escherichia coli isolated from different organs of the individual sheep: molecular typing, antibiotics resistance, and biofilm formation.

Despite numerous studies on Escherichia coli (E. coli) from sheep, there have been few reports on the characterization of E. coli isolates from various organs of individual sheep until now. The present study conducted molecular typing, antibiotics resistance, biofilm formation, and virulence genes on E. coli isolated from 57 freshly slaughtered apparently healthy sheep carcasses, gallbladders, fecal samples, and mesenteric lymph nodes (MLNs). The results demonstrated that the detection rate of R1 LPS core type in E. coli isolated from fecal samples (70.83%) was higher than that from other organs, but the detection rate of antibiotic resistance genes was lower (P < 0.05). The predominant phylogenetic group of E. coli isolated from the carcasses was group B1 (93.33%), and the detection rate of multidrug-resistance phenotype (80%) and the resistance rate of E. coli was higher than that from other organs (P < 0.05). Interestingly, the intensity of biofilm formation of E. coli isolated from MLNs was higher than that from other organs (P < 0.05). However, except for ibeB, the detection rates of virulence genes did not differ in E.coli isolated from different organs. In conclusion, differences were noted in these parameters of E. coli isolated from different organs of individual sheep. Therefore, the data may contain considerable mistakes concerning the actual situation in the host if we only analyze the data of E. coli isolated from feces or carcasses.

In situanchoring of Fe3O4/CNTs hybrids on basalt fiber for efficient electromagnetic wave absorption.

The development of practical and efficient electromagnetic wave (EMW) absorbing materials is a challenging research problem. A mussel-inspired molecular structure regulation strategy using polydopamine to increase the roughness and functional groups of basalt fiber (BF) surface, which can improve the fiber interfacial adhesion. Herein, a novel BF-Fe3O4/CNTs heterostructure is synthesized through a dip-coating adsorption process. The three-dimensional network structure of Fe3O4/CNTs hybridin situanchored on the surface of BF, which endows the composite to have good intrinsic magnetic and dielectric properties. Modulation of EMW absorption performance by controlling the addition of CNTs, the minimum RL of BF-Fe3O4/7C reaches to -40.57 dB at a thickness of 1.5 mm with CNTs addition of 7%. The enhanced EMW absorption performance of BF-Fe3O4/7C heterostructure may be attributed to the synergistic effects of interfacial polarization between the hollow magnetic Fe3O4spheres and CNTs, conduction loss, magnetic resonance loss and multiple reflection/scattering inside the BF. This work provides a simple pathway to design EMW absorbing materials with good environmental stability.

A generic framework for embedding human brain function with temporally correlated autoencoder.

Learning an effective and compact representation of human brain function from high-dimensional fMRI data is crucial for studying the brain's functional organization. Traditional representation methods such as independent component analysis (ICA) and sparse dictionary learning (SDL) mainly rely on matrix decomposition which represents the brain function as spatial brain networks and the corresponding temporal patterns. The correspondence of those brain networks across individuals are built by viewing them as one-hot vectors and then performing the matching. However, those one-hot vectors do not encode the regularity and/or variability of different brains very well, and thus are limited in effectively representing the functional brain activities across individuals and among different time points. To address this problem, in this paper, we formulate the human brain functional representation as an embedding problem, and propose a novel embedding framework based on the Transformer model to encode the brain function in a compact, stereotyped and comparable latent space where the brain activities are represented as dense embedding vectors. We evaluate the proposed embedding framework on the publicly available Human Connectome Project (HCP) task fMRI dataset. The experiments on brain state prediction task indicate the effectiveness and generalizability of the learned embedding. We also explore the interpretability of the learned embedding from both spatial and temporal perspective. In general, our approach provides novel insights on representing the regularity and variability of human brain function in a general, comparable, and stereotyped latent space.

SNP application improves drought tolerance in soybean.

As an important bioactive molecule, nitric oxide (NO) can effectively alleviate the effects of drought stress on crops. However, it is still unclear whether it can increase the stress resistance of soybean. Therefore, in this study, our objective was to explore the effect of exogenous NO application on the physiological characteristics of soybean seedlings under drought stress. As test material, two soybean varieties, HN65 and HN44, were used, while sodium nitroprusside (SNP) of 100 µmol L-1, 200 µmol L-1, 500 µmol L-1, 1000 µmol L-1 served as an exogenous NO donor, and PEG-6000 as an osmotic regulator to simulate drought stress. The effects of irrigation with different SNP concentrations for different days on the physiological characteristics of the soybean seedlings under drought conditions were then investigated. The results obtained showed that the activities of antioxidant enzymes, osmotic regulator contents, as well as the abscisic acid and salicylic acid contents of the plant leaves increased with increasing SNP concentration and treatment time. However, we observed that excessively high SNP concentrations decreased the activities of key nitrogen metabolism enzymes significantly. This study provides a theoretical basis for determining a suitable exogenous NO concentration and application duration. It also highlights strategies for exploring the mechanism by which exogenous NO regulates crop drought resistance.

A deep learning method for autism spectrum disorder identification based on interactions of hierarchical brain networks.

BACKGROUND: It has been recently shown that deep learning models exhibited remarkable performance of representing functional Magnetic Resonance Imaging (fMRI) data for the understanding of brain functional activities. With hierarchical structure, deep learning models can infer hierarchical functional brain networks (FBN) from fMRI. However, the applications of the hierarchical FBNs have been rarely studied. METHODS: In this work, we proposed a hierarchical recurrent variational auto-encoder (HRVAE) to unsupervisedly model the fMRI data. The trained HRVAE encoder can predict hierarchical temporal features from its three hidden layers, and thus can be regarded as a hierarchical feature extractor. Then LASSO (least absolute shrinkage and selection operator) regression was applied to estimate the corresponding hierarchical FBNs. Based on the hierarchical FBNs from each subject, we constructed a novel classification framework for brain disorder identification and test it on the Autism Brain Imaging Data Exchange (ABIDE) dataset, a world-wide multi-site database of autism spectrum disorder (ASD). We analyzed the hierarchy organization of FBNs, and finally used the overlaps of hierarchical FBNs as features to differentiate ASD from typically developing controls (TDC). RESULTS: The experimental results on 871 subjects from ABIDE dataset showed that the HRVAE model can effectively derive hierarchical FBNs including many well-known resting state networks (RSN). Moreover, the classification result improved the state-of-the-art by achieving a very high accuracy of 82.1 %. CONCLUSIONS: This work presents a novel data-driven deep learning method using fMRI data for ASD identification, which could provide valuable reference for clinical diagnosis. The classification results suggest that the interactions of hierarchical FBNs have association with brain disorder, which promotes the understanding of FBN hierarchy and could be applied to other brain disorder analysis.

Microwell Confined Electro-Coalescence for Rapid Formation of High-Throughput Droplet Array.

Droplet array is widely applied in single cell analysis, drug screening, protein crystallization, etc. This work proposes and validates a method for rapid formation of uniform droplet array based on microwell confined droplets electro-coalescence of screen-printed emulsion droplets, namely electro-coalescence droplet array (ECDA). The electro-coalescence of droplets is according to the polarization induced electrostatic and dielectrophoretic forces, and the dielectrowetting effect. The photolithographically fabricated microwells are highly regular and reproducible, ensuring identical volume and physical confinement to achieve uniform droplet array, and meanwhile the microwell isolation protects the paired water droplets from further fusion and broadens its feasibility to different fluidic systems. Under optimized conditions, a droplet array with an average diameter of 85 µm and a throughput of 106 in a 10 cm × 10 cm chip can be achieved within 5 s at 120 Vpp and 50 kHz. This ECDA chip is validated for various microwell geometries and functional materials. The optimized ECDA are successfully applied for digital viable bacteria counting, showing comparable results to the plate culture counting. Such an ECDA chip, as a digitizable and high-throughput platform, presents excellent potential for high-throughput screening, analysis, absolute quantification, etc.

Microstructure, Mechanical and Tribological Properties of High-Entropy Carbide (MoNbTaTiV)C5.

High-entropy carbide (NbTaTiV)C4 (HEC4), (MoNbTaTiV)C5 (HEC5), and (MoNbTaTiV)C5-SiC (HEC5S) multiphase ceramics were prepared by spark plasma sintering (SPS) at 1900 to 2100 °C, using metal carbide and silicon carbide (SiC) as raw materials. Their microstructure, and mechanical and tribological properties were investigated. The results showed that the (MoNbTaTiV)C5 synthesized at 1900-2100 °C had a face-centered cubic structure and density higher than 95.6%. The increase in sintering temperature was conducive to the promotion of densification, growth of grains, and diffusion of metal elements. The introduction of SiC helped to promote densification but weakened the strength of the grain boundaries. The average specific wear rates for HEC4 were within an order of magnitude of 10-5 mm3/N·m, and for HEC5 and HEC5S were within a range of 10-7 to 10-6 mm3/N·m. The wear mechanism of HEC4 was abrasion, while that of HEC5 and HEC5S was mainly oxidation wear.