Showing papers by "Jiawei Han published in 2023"
TL;DR: This article proposed a prompt-enhanced diffusion model that alleviates data scarcity with orders of magnitude concept compositions by using language-free audios and leveraged spectrogram autoencoder to predict the self-supervised audio representation instead of waveforms.
Abstract: Large-scale multimodal generative modeling has created milestones in text-to-image and text-to-video generation. Its application to audio still lags behind for two main reasons: the lack of large-scale datasets with high-quality text-audio pairs, and the complexity of modeling long continuous audio data. In this work, we propose Make-An-Audio with a prompt-enhanced diffusion model that addresses these gaps by 1) introducing pseudo prompt enhancement with a distill-then-reprogram approach, it alleviates data scarcity with orders of magnitude concept compositions by using language-free audios; 2) leveraging spectrogram autoencoder to predict the self-supervised audio representation instead of waveforms. Together with robust contrastive language-audio pretraining (CLAP) representations, Make-An-Audio achieves state-of-the-art results in both objective and subjective benchmark evaluation. Moreover, we present its controllability and generalization for X-to-Audio with"No Modality Left Behind", for the first time unlocking the ability to generate high-definition, high-fidelity audios given a user-defined modality input. Audio samples are available at https://Text-to-Audio.github.io
33 citations
TL;DR: AIGC-Audio as mentioned in this paper proposes a multi-modal AI system named AudioGPT, which complements ChatGPT with foundation models to process complex audio information and solve numerous understanding and generation tasks.
Abstract: Large language models (LLMs) have exhibited remarkable capabilities across a variety of domains and tasks, challenging our understanding of learning and cognition. Despite the recent success, current LLMs are not capable of processing complex audio information or conducting spoken conversations (like Siri or Alexa). In this work, we propose a multi-modal AI system named AudioGPT, which complements LLMs (i.e., ChatGPT) with 1) foundation models to process complex audio information and solve numerous understanding and generation tasks; and 2) the input/output interface (ASR, TTS) to support spoken dialogue. With an increasing demand to evaluate multi-modal LLMs of human intention understanding and cooperation with foundation models, we outline the principles and processes and test AudioGPT in terms of consistency, capability, and robustness. Experimental results demonstrate the capabilities of AudioGPT in solving AI tasks with speech, music, sound, and talking head understanding and generation in multi-round dialogues, which empower humans to create rich and diverse audio content with unprecedented ease. Our system is publicly available at \url{https://github.com/AIGC-Audio/AudioGPT}.
21 citations
TL;DR: In this article , the authors provide a comprehensive discussion of challenges and opportunities for the three core components of the treatment effect estimation task, i.e., treatment, covariates, and outcome.
Abstract: Causal inference has numerous real-world applications in many domains, such as health care, marketing, political science, and online advertising. Treatment effect estimation, a fundamental problem in causal inference, has been extensively studied in statistics for decades. However, traditional treatment effect estimation methods may not well handle large-scale and high-dimensional heterogeneous data. In recent years, an emerging research direction has attracted increasing attention in the broad artificial intelligence field, which combines the advantages of traditional treatment effect estimation approaches (e.g., propensity score, matching, and reweighing) and advanced machine learning approaches (e.g., representation learning, adversarial learning, and graph neural networks). Although the advanced machine learning approaches have shown extraordinary performance in treatment effect estimation, it also comes with a lot of new topics and new research questions. In view of the latest research efforts in the causal inference field, we provide a comprehensive discussion of challenges and opportunities for the three core components of the treatment effect estimation task, i.e., treatment, covariates, and outcome. In addition, we showcase the promising research directions of this topic from multiple perspectives.
4 citations
TL;DR: In this paper , the performance of three voltage control strategies for DC microgrids are compared, including the proportion integration (PI) control, the fuzzy PI control and particle swarm optimization (PSO) PI control.
Abstract: Direct-current (DC) microgrids have gained worldwide attention in recent decades due to their high system efficiency and simple control. In a self-sufficient energy system, voltage control is an important key to dealing with upcoming challenges of renewable energy integration into DC microgrids, and thus energy storage systems (ESSs) are often employed to suppress the power fluctuation and ensure the voltage stability. In this paper, the performances of three voltage control strategies for DC microgrids are compared, including the proportion integration (PI) control, the fuzzy PI control and particle swarm optimization (PSO) PI control. Particularly, two kinds of ESSs including battery and advanced adiabatic compressed air energy storage (AA-CAES) with different operational characteristics are installed in the microgrid, and their impacts on voltage control are investigated. The control performances are comprehensively compared under different control schemes, various scenarios of renewable energy fluctuations, participation in the control of the two ESSs or not, and different fault conditions. Additionally, the dynamic performances of the ESSs are exhibited. The results verify the validity of the control schemes and the feasibility of the configuration of the ESSs into the DC microgrid.
2 citations
TL;DR: In this paper , the pitch contour was used as an auxiliary feature and introduced a temporal loss in the facial motion prediction process to regulate the outliers in the predicted motion sequence to avoid robustness issues.
Abstract: Generating talking person portraits with arbitrary speech audio is a crucial problem in the field of digital human and metaverse. A modern talking face generation method is expected to achieve the goals of generalized audio-lip synchronization, good video quality, and high system efficiency. Recently, neural radiance field (NeRF) has become a popular rendering technique in this field since it could achieve high-fidelity and 3D-consistent talking face generation with a few-minute-long training video. However, there still exist several challenges for NeRF-based methods: 1) as for the lip synchronization, it is hard to generate a long facial motion sequence of high temporal consistency and audio-lip accuracy; 2) as for the video quality, due to the limited data used to train the renderer, it is vulnerable to out-of-domain input condition and produce bad rendering results occasionally; 3) as for the system efficiency, the slow training and inference speed of the vanilla NeRF severely obstruct its usage in real-world applications. In this paper, we propose GeneFace++ to handle these challenges by 1) utilizing the pitch contour as an auxiliary feature and introducing a temporal loss in the facial motion prediction process; 2) proposing a landmark locally linear embedding method to regulate the outliers in the predicted motion sequence to avoid robustness issues; 3) designing a computationally efficient NeRF-based motion-to-video renderer to achieves fast training and real-time inference. With these settings, GeneFace++ becomes the first NeRF-based method that achieves stable and real-time talking face generation with generalized audio-lip synchronization. Extensive experiments show that our method outperforms state-of-the-art baselines in terms of subjective and objective evaluation. Video samples are available at https://genefaceplusplus.github.io .
1 citations
TL;DR: In this paper , a mini-review summarizes the main progress and related regulatory mechanisms of Na+ uptake in teleost ionocytes, and discusses some of the challenges to the current models.
Abstract: How teleosts take up Na+ from the surrounding freshwater (FW) as well as the underlying mechanisms associated with this process have received considerable attention over the past 85 years. Owing to an enormous ion gradient between hypotonic FW and fish body fluids, teleosts gills have to actively absorb Na+ (via ionocytes) to compensate for the passive loss of Na+. To date, three models have been proposed for Na+ uptake in teleost ionocytes, including Na+/H+ exchanger (NHE)-mediated, acid-sensing ion channel (ASIC)-mediated, Na+-Cl- co-transporter (NCC)-mediated pathways. However, some debates regarding these models and unclear mechanisms still remain. To better understand how teleosts take up Na+ from FW, this mini-review summarizes the main progress and related regulatory mechanisms of Na+ uptake, and discusses some of the challenges to the current models.
1 citations
TL;DR: In this paper , a single-story full-scale timber structure with four columns and Dou-gongs was taken as the object, and pseudo-static test (or simulation) analysis was carried out on the FTM and the corresponding FEM.
Abstract: In order to accurately use numerical analysis results to evaluate seismic performance, a quantitative relationship between the parameters of the ideal finite element model (FEM) and full-scale test model (FTM) of historic timber structures must be established. In this study, a single-story full-scale timber structure with four columns and Dou-gongs that was common during the Song dynasty of China (A.D. 960–1279) was taken as the object, and pseudo-static test (or simulation) analysis was carried out on the FTM and the corresponding FEM. The numerical model was consistent with the test model in terms of the structural deformation pattern and stress characteristics of the key nodes. The key-point loads of the skeleton curves, stiffness ratio, ductility coefficient, and stiffness relationship were quantitatively established for the two models. Accordingly, approximate formulas for the above relationships were proposed.
TL;DR: A simple and convenient electrochemical sensor for Lysozyme detection was prepared by modifying gold nanoparticles (AuNPs) and ferrocene dicarboxylic acid (Fc(COOH)2) on a glass carbon electrode (GCE), which was characterized fully by various electrochemical methods and field emission scanning electron microscope (FESEM) as discussed by the authors .
Abstract: Lysozyme (Lyz) is found in animal and human bodily fluids, and is frequently utilized as a biomarker for various diseases. Even trace amounts of Lyz in food can potentially trigger adverse immune system reactions in sensitive individuals. Therefore, it is very important to monitor Lyz concentration in foods for safety. In this study, a simple and convenient electrochemical sensor for Lyz detection was prepared by modifying gold nanoparticles (AuNPs) and ferrocene dicarboxylic acid (Fc(COOH)2) on a glass carbon electrode (GCE), which was characterized fully by various electrochemical methods and field emission scanning electron microscope (FESEM). The proposed method utilized Fc(COOH)2 as a probe and AuNPs as an electron transfer medium to improve the sensor’s current response performance. Under optimal conditions, the sensor was used to detect Lyz with a linear range from 0.10~0.70 mmol·L−1 with a sensitivity of 50.55 μA·mM−1·cm−2, and a limit of detection (LOD) of 0.07 mmol·L−1. In the standard addition experiment of food samples (egg white), a total R.S.D. of less than 6.75% and an average recovery between 95.45% and 102.62% were obtained.
TL;DR: In this paper , depth odometry is achieved only by using the depth information from the depth camera, and the point cloud cross-source map registration is realized by 3D particle filtering to obtain the pose of the point clouds relative to the map.
Abstract: In this study, a localisation system without cumulative errors is proposed. First, depth odometry is achieved only by using the depth information from the depth camera. Then the point cloud cross-source map registration is realised by 3D particle filtering to obtain the pose of the point cloud relative to the map. Furthermore, we fuse the odometry results with the point cloud to map registration results, so the system can operate effectively even if the map is incomplete. The effectiveness of the system for long-term localisation, localisation in the incomplete map, and localisation in low light through multiple experiments on the self-recorded dataset is demonstrated. Compared with other methods, the results are better than theirs and achieve high indoor localisation accuracy.
TL;DR: In this paper , the effects of trace Si element on the microstructure, mechanical properties and high temperature softening resistance of Cu-0.9Ni-1.1 P (mass fraction, %) alloy were studied using hardness tests, conductivity measurements, SEM, TEM and XRD observations.
Abstract: The effects of trace Si element on the microstructure, mechanical properties and high temperature softening resistance of Cu-0.9Ni-1.9Sn-0.1 P (mass fraction, %) alloy were studied using hardness tests, conductivity measurements, SEM, TEM and XRD observations. According to the results, after 70% cold rolling and aging at 450 ℃, the micro-hardness, conductivity and softening temperatures of Cu-0.9Ni-1.9Sn-0.1 P-0.1Si alloy were up to 224.2 HV, 36.4%IACS and 520 ℃. In comparison with Cu-0.9Ni-1.9Sn-0.1 P alloy, the micro-hardness of Cu-0.9Ni-1.9Sn-0.1 P-0.1Si alloy was increased from 192.3 HV to 224.2 HV, and the conductivity increased from 33.6%IACS to 36.4%IACS. TEM analysis shows that adding Si can from precipitated Ni2Si phase with Ni and increase the strength of Cu-0.9Ni-1.9Sn-0.1 P-0.1Si alloy. Moreover, the addition of 0.1% Si can increase the softening temperature of the Cu-0.9Ni-1.9Sn-0.1 P alloy. As a result of annealing at different temperatures, the softening mechanism of the alloy varies. When the annealing at low temperature ( 400 ℃−480 ℃), the annealing softening is related to recovery and recrystallization. When the annealing temperature is high ( 480 ℃−580 ℃), the annealing softening is primarily caused by recrystallized grain growth and the coarsening of second-phase particles．
TL;DR: This article proposed a latent diffusion-based T2A method that uses pre-trained large language models (LLMs) to parse the text into structuredpairs for better temporal information capture and introduce another structured-text encoder to aid in learning semantic alignment during the diffusion denoising process.
Abstract: Large diffusion models have been successful in text-to-audio (T2A) synthesis tasks, but they often suffer from common issues such as semantic misalignment and poor temporal consistency due to limited natural language understanding and data scarcity. Additionally, 2D spatial structures widely used in T2A works lead to unsatisfactory audio quality when generating variable-length audio samples since they do not adequately prioritize temporal information. To address these challenges, we propose Make-an-Audio 2, a latent diffusion-based T2A method that builds on the success of Make-an-Audio. Our approach includes several techniques to improve semantic alignment and temporal consistency: Firstly, we use pre-trained large language models (LLMs) to parse the text into structuredpairs for better temporal information capture. We also introduce another structured-text encoder to aid in learning semantic alignment during the diffusion denoising process. To improve the performance of variable length generation and enhance the temporal information extraction, we design a feed-forward Transformer-based diffusion denoiser. Finally, we use LLMs to augment and transform a large amount of audio-label data into audio-text datasets to alleviate the problem of scarcity of temporal data. Extensive experiments show that our method outperforms baseline models in both objective and subjective metrics, and achieves significant gains in temporal information understanding, semantic consistency, and sound quality.
TL;DR: In this paper , the difference in immunomodulatory effects between collagen-derived dipeptides and amino acids was investigated, and the authors concluded that there was no difference in cytokine secretion between amino acids and their respective amino acids.
Abstract: A number of food components, such as polyphenols and phytonutrients, have immunomodulatory effects. Collagen has various bioactivities, such as antioxidative effects, the promotion of wound healing, and relieving symptoms of bone/joint disease. Collagen is digested into dipeptides and amino acids in the gastrointestinal tract and subsequently absorbed. However, the difference in immunomodulatory effects between collagen-derived dipeptides and amino acids is unknown. To investigate such differences, we incubated M1 macrophages or peripheral blood mononuclear cells (PBMC) with collagen-derived dipeptides (hydroxyproline-glycine (Hyp-Gly) and proline-hydroxyproline (Pro-Hyp)) and amino acids (proline (Pro), hydroxyproline (Hyp), and glycine (Gly)). We first investigated the dose dependency of Hyp-Gly on cytokine secretion. Hyp-Gly modulates cytokine secretion from M1 macrophages at 100 µM, but not at 10 µM and 1 µM. We then compared immunomodulatory effects between dipeptides and mixtures of amino acids on M1 macrophages and PBMC. There was, however, no difference in cytokine secretion between dipeptides and their respective amino acids. We conclude that collagen-derived dipeptides and amino acids have immunomodulatory effects on M1-differentiated RAW264.7 cells and PBMC and that there is no difference in the immunomodulatory effects between dipeptides and amino acids.
TL;DR: In this paper , the authors performed a genome-wide association study (GWAS) on C. acnes and found that all subspecies and phylotypes of the organism, possibly with the exception of C. elongatum, are able to cause deep-seated infection given favorable conditions, most importantly related to inserted foreign material.
Abstract: Opportunistic infections emerging from human skin microbiota are of ever-increasing importance. Cutibacterium acnes, being abundant on the human skin, may cause deep-seated infections (e.g., device-associated infections). Differentiation between invasive (i.e., clinically significant) C. acnes isolates and sole contaminants is often difficult. ABSTRACT Cutibacterium acnes, formerly known as Propionibacterium acnes, is a commensal of the human pilosebaceous unit but also causes deep-seated infection, especially in the context of orthopedic and neurosurgical foreign materials. Interestingly, little is known about the role of specific pathogenicity factors for infection establishment. Here, 86 infection-associated and 103 commensalism-associated isolates of C. acnes were collected from three independent microbiology laboratories. We sequenced the whole genomes of the isolates for genotyping and a genome-wide association study (GWAS). We found that C. acnes subsp. acnes IA1 was the most significant phylotype among the infection isolates (48.3% of all infection isolates; odds ratio [OR] = 1.98 for infection). Among the commensal isolates, C. acnes subsp. acnes IB was the most significant phylotype (40.8% of all commensal isolates; OR = 0.5 for infection). Interestingly, C. acnes subsp. elongatum (III) was rare overall and did not occur at all in infection. The open reading frame-based GWAS (ORF-GWAS) did not show any loci with a strong signal for infection association (no P values of ≤0.05 after adjustment for multiple testing; no logarithmic OR [logOR] of ≥|2|). We concluded that all subspecies and phylotypes of C. acnes, possibly with the exception of C. acnes subsp. elongatum, are able to cause deep-seated infection given favorable conditions, most importantly related to inserted foreign material. Genetic content appears to have a small effect on the likelihood of infection establishment, and functional studies are needed to understand the individual factors contributing to deep-seated infections caused by C. acnes. IMPORTANCE Opportunistic infections emerging from human skin microbiota are of ever-increasing importance. Cutibacterium acnes, being abundant on the human skin, may cause deep-seated infections (e.g., device-associated infections). Differentiation between invasive (i.e., clinically significant) C. acnes isolates and sole contaminants is often difficult. Identification of genetic markers associated with invasiveness not only would strengthen our knowledge related to pathogenesis but also could open ways to selectively categorize invasive and contaminating isolates in the clinical microbiology lab. We show that in contrast to other opportunistic pathogens (e.g., Staphylococcus epidermidis), invasiveness is apparently a broadly distributed ability across almost all C. acnes subspecies and phylotypes. Thus, our work strongly supports an approach in which clinical significance is judged from clinical context rather than by detecting specific genetic traits.
TL;DR: In this article , the authors aimed at the stick-slip portion of the creeping Qianning segment of the Xianshuihe fault to determine the characteristics of fault rocks and how fluids at depth influence fault behavior.
Abstract: While the Xianshuihe fault displays continuous creeping behavior, it is also the most seismically active fault in the eastern Tibetan Plateau, and its earthquake mechanisms remain unclear. Here, we aim at the stick-slip portion of the creeping Qianning segment of the Xianshuihe fault to determine the characteristics of fault rocks and how fluids at depth influence fault behavior. Field survey, optical and scanning electron microscope observations, X-ray diffraction and fluorescence analyses, as well as carbon and oxygen isotope analyses were performed on the collected samples. The fault core consists of 3 to 5 cm-thick black fault gouge and ∼2.5 m-thick breccia, surrounded by ∼12 m damage zone. In contrast to fault breccia (1 to 8 cm in diameter), the black fault gouge, which represents the principal slip zone of repeated seismic events, contains angular quartz particles (∼10 μm on average) and clays dominated by illite. The fluid-rock interactions altering silica minerals into illite, and the thermal decomposition of carbonate minerals, passively increase the relative content of quartz and feldspar (total 63-73%) in the fault gouge. The deeply-sourced CO2 (from mantle and metamorphic degassing) within the hydrothermal fluids causes carbonate precipitation in breccias (21-53%), composed of calcite, dolomite, and aragonite. These fluid-assisted reactions lead to more abundant strong mineral phases (quartz, feldspar, and carbonates, 64-87%) than weak clays (12-36%) within the fault core, and locally strengthen the fault, which inhibits slow release of stress at shallow depth and promotes seismic rupture of the fault.
TL;DR: FusionTrack as mentioned in this paper utilizes a joint track-detection decoder and a score-guided multi-level query fuser to enhance the usage of information within and between frames.
Abstract: Multi-object tracking (MOT) is one of the significant directions of computer vision. Though existing methods can solve simple tasks like pedestrian tracking well, some complex downstream tasks featuring uniform appearance and diverse motion remain difficult. Inspired by DETR, the tracking-by-attention (TBA) method uses transformers to accomplish multi-object tracking tasks. However, there are still issues with existing TBA methods within the TBA paradigm, such as difficulty detecting and tracking objects due to gradient conflict in shared parameters, and insufficient use of features to distinguish similar objects. We introduce FusionTrack to address these issues. It utilizes a joint track-detection decoder and a score-guided multi-level query fuser to enhance the usage of information within and between frames. With these improvements, FusionTrack achieves 11.1% higher by HOTA metric on the DanceTrack dataset compared with the baseline model MOTR.
TL;DR: In this paper , the authors document the past 1.1 My of contourite succession at IODP Site U1389, in which there are a total of 299 full and partial sequences, with a variable thickness of 0.13-10.6 m (mean 2.65 m).
TL;DR: In this paper , the authors explored the paleo-drainage pattern of the Dadu and Anning Rivers based on bulk mineral and geochemical analyses of the large quantities of fluvial/lacustrine sediments along the trunk of theDadu River.
Abstract: The Xianshuihe-Anninghe fault extents SE–S and constitutes the southeastern margin of the Tibetan Plateau. However, the associated Dadu River does not flow following the fault, but makes a 90º turn within a distance of 1 km at Shimian, heading east, and joins the Yangtze River, finally flowing into the East China Sea. Adjacent to the abrupt turn, a low and wide pass near the Daqiao reservoir at Mianning separates the N–S course of the Dadu River from the headwater of the Anning River which then flows south into the Yunnan Province along the Anninghe fault. Therefore, many previous studies assumed southward flow of the paleo-Dadu River from the Shimian to the Anning River. However, evidences for the capture of the integrated N–S paleo-Dadu-Anning River, its timing, and causes are still insufficient. This study explored the paleo-drainage pattern of the Dadu and Anning Rivers based on bulk mineral and geochemical analyses of the large quantities of fluvial/lacustrine sediments along the trunk of the Dadu and Anning Rivers. Similar with sands in the modern Dadu River, the Xigeda sediments also exhibit a granitoid affinity with the bulk major mineral compositions of quartz (>50%), anorthite (about 10%), orthoclase (about 5%), muscovite (about 5%), and clinochlore (about 4%). Correspondingly, bulk major elements show high SiO2, with all samples >60%, and some of them >70%, low TiO2 (≤0.75%), P2O5 (≤0.55%), FeO* (≤5%), and relatively high CaO (1.02%–8.51%), Na2O (1.60%–2.52%), and K2O (2.17%–2.71%), with a uniform REE patterns. Therefore, synthesizing all these results indicate that these lacustrine sediments have similar material sources, which are mainly derived from its course in the Songpan-Ganzi flysch block, implying that the paleo-Dadu originally flowed southward into the Anning River and provided materials to the Xigeda ancient lake. The rearrangement of the paleo-Dadu River appears to be closely related to the locally focused uplift driven by strong activities of the Xianshuihe-Xiaojiang fault system.
TL;DR: In this article , the reliability of the complex electronic components for airborne applications under a thermal cycling test, random vibration and combined loading has been investigated by experiment tests and finite element simulation, and the results indicated that the combined fatigue life was much shorter than a single-factor experiment.
Abstract: The electronic devices suffer great vibration and temperature fluctuation in an airborne environment, which has been always a big challenge for reliability design. In this paper, the reliability of the complex electronic components for airborne applications under a thermal cycling test, random vibration and combined loading has been investigated by experiment tests and finite element simulation. The fatigue life and failure location under different loadings have been compared and discussed, respectively. The results indicated that the combined fatigue life was much shorter than a single-factor experiment. The failed solder joints mostly appeared at the interface between the solder and the copper pad on the component side and the location was at the corner for all three harsh environment tests. Nevertheless, several differences could be observed. For temperature cycling, all the specimens failed due to the increase in daisy chain resistance rather than the open circuit for the combined loading test. That is because the degeneration of the solder caused by temperature variation led to lower stress levels and fatigue life. Moreover, the pins fractured at the welding regions have been observed. The modified Coffin—Manson model, Miner’s linear fatigue damage criterion and Steinberg’s model and rapid life-prediction approach were used to predict the fatigue life under temperature cycling, random vibration and combined loading, respectively. With these methods, the accurate numerical models could be developed and validated by experiment results. Thanks to the simulation, the design time could be effectively shortened and the weak point could be determined.
TL;DR: In this article , a nanomechanical biosensor based on Fabry-Pérot interference demodulation was developed for antibiotic susceptibility testing (AST) to inhibit the overuse of antibiotics.
Abstract: There is an urgent need for developing rapid and affordable antibiotic susceptibility testing (AST) technologies to inhibit the overuse of antibiotics. In this study, a novel microcantilever nanomechanical biosensor based on Fabry-Pérot interference demodulation was developed for AST. To construct the biosensor, a cantilever was integrated with the single mode fiber in order to form the Fabry-Pérot interferometer (FPI). After the attachment of bacteria on the cantilever, the fluctuations of cantilever caused by the bacterial movements were detected by monitoring the changes of resonance wavelength in the interference spectrum. We applied this methodology to Escherichia coli and Staphylococcus aureus, showing the amplitude of cantilever's fluctuations was positively related on the quantity of bacteria immobilized on the cantilever and associated with the bacterial metabolism. The response of bacteria to antibiotics was dependent on the types of bacteria, the types and concentrations of antibiotics. Moreover, the minimum inhibitory and bactericidal concentrations for Escherichia coli were obtained within 30 minutes, demonstrating the capacity of this method for rapid AST. Benefiting from the simplicity and portability of the optical fiber FPI-based nanomotion detection device, the developed nanomechanical biosensor in this study provides a promising technique for AST and a more rapid alternative for clinical laboratories.
04 Jun 2023
TL;DR: In this article , a weakly-supervised method with real-synthetic hybrid data is proposed, which only requires a small portion of unlabeled real images and auto-generated synthetic labeled images for training.
Abstract: Due to the domain gap between the public large-scale datasets and actual scenes, the crowd counting models trained on the common datasets have a significant performance degradation when applying in practical applications. To address the above issue, one of the solution is to label additional data from the novel scenes, which is time-consuming and impractical for multiple scenes. Another solution is to utilize domain adaptation approaches to adapt a well-trained model to novel scenes. However, most of these approaches focus on appearance adaptation while the background and the crowd distribution is not adapted. In this paper, we propose a weakly-supervised method with real-synthetic hybrid data which only requires a small portion of unlabelled real images and auto-generated synthetic labelled images for training. First, the hybrid data is generated based on background from the real scene and random distributed synthetic persons. Second, an initialized counter is trained based on the hybrid data and the crowd distribution is predicted based on the predictions on real images. Then, a better crowd counter is trained based on new hybrid data generated from updated crowd distribution. The process is iterated until convergence. Extensive experiments demonstrate the effectiveness of the proposed method.
TL;DR: Zhang et al. as mentioned in this paper proposed a many-objective evolutionary optimization algorithm based on indicator and decomposition (IDEA) to keep the convergence and diversity simultaneously, which is very effective compared to ten state-of-the-art manyobjective algorithms.
Abstract: In the field of many-objective evolutionary optimization algorithms (MaOEAs), how to maintain the balance between convergence and diversity has been a significant research problem. With the increase of the number of objectives, the number of mutually nondominated solutions increases rapidly, and multi-objective evolutionary optimization algorithms, based on Pareto-dominated relations, become invalid because of the loss of selection pressure in environmental selection. In order to solve this problem, indicator-based many-objective evolutionary algorithms have been proposed; however, they are not good enough at maintaining diversity. Decomposition-based methods have achieved promising performance in keeping diversity. In this paper, we propose a MaOEA based on indicator and decomposition (IDEA) to keep the convergence and diversity simultaneously. Moreover, decomposition-based algorithms do not work well on irregular PFs. To tackle this problem, this paper develops a reference-points adjustment method based on the learning population. Experimental studies of several well-known benchmark problems show that IDEA is very effective compared to ten state-of-the-art many-objective algorithms.
TL;DR: In this paper , the authors review the recent advances in pathophysiology management following a traumatic hemorrhage as well as the role of diagnostic imaging in identifying the source of hemorrhage.
Abstract: Abstract Trauma is the number one cause of death among Americans between the ages of 1 and 46 years, costing more than $670 billion a year. Following death related to central nervous system injury, hemorrhage accounts for the majority of remaining traumatic fatalities. Among those with severe trauma that reach the hospital alive, many may survive if the hemorrhage and traumatic injuries are diagnosed and adequately treated in a timely fashion. This article aims to review the recent advances in pathophysiology management following a traumatic hemorrhage as well as the role of diagnostic imaging in identifying the source of hemorrhage. The principles of damage control resuscitation and damage control surgery are also discussed. The chain of survival for severe hemorrhage begins with primary prevention; however, once trauma has occurred, prehospital interventions and hospital care with early injury recognition, resuscitation, definitive hemostasis, and achieving endpoints of resuscitation become paramount. An algorithm is proposed for achieving these goals in a timely fashion as the median time from onset of hemorrhagic shock and death is 2 h.
TL;DR: In this paper , it was shown that the multifunctional Yersinia effector YopM inhibits effector triggered immunity and increases production of the anti-inflammatory cytokine Interleukin-10 (IL-10) to suppress the host immune response.
Abstract: The multifunctional Yersinia effector YopM inhibits effector triggered immunity and increases production of the anti-inflammatory cytokine Interleukin-10 (IL-10) to suppress the host immune response. Previously it was shown that YopM induces IL-10 gene expression by elevating phosphorylation of the serine-threonine kinase RSK1 in the nucleus of human macrophages. Using transcriptomics, we now show that YopM affects expression of genes encoding components of the JAK-STAT signaling pathway. Further analysis revealed that YopM mediates nuclear translocation of the transcription factor Stat3 in Y. enterocolitica infected macrophages and that knockdown of Stat3 inhibited YopM-induced IL-10 gene expression. YopM-induced Stat3 translocation did not depend on autocrine IL-10, activation of RSK1 or tyrosine phosphorylation of Stat3. Thus, besides activation of RSK1, stimulation of nuclear translocation of Stat3 is another mechanism by which YopM increases IL-10 gene expression in macrophages.
01 Jan 2023
TL;DR: Based on literature review and analyses, this article provided some updated information regarding the epidemiological characteristics of Omicron infection and the impact of infection on physiological functions, and also offered some suggestions on the surgical management of omicron patients.
Abstract:
The Omicron strain, the new COVID-19 variant prevalent in China currently, was characterized by higher infectivity and stronger immune evasion than previous strains of COVID-19. Existing research evidence showed that Omicron causes serious disturbances to normal physiological functions and potentially increases the perioperative risk to surgical patients. Understanding the epidemiological characteristics and the influence of Omicron on physiological functions would help selecting the most appropriate timing of surgery, which could be critical to the surgical outcomes of these Omicron patients. Based on literature review and analyses, this article would offer some updated information regarding the epidemiological characteristics of Omicron infection and the impact of Omicron infection on physiological functions, and also offer some suggestions on the surgical management of Omicron patients.
TL;DR: The cosmogenic nuclide burial ages of the lacustrine sediments along the Dadu and Anning rivers suggest deposition of these sediments from separate dammed lakes ca. 1.2 Ma, ca. 0.6 Ma, and ca.0.9 Ma, respectively as mentioned in this paper .
TL;DR: The relationship between metal exposure/essential metal dyshomeostasis and Pulmonary arterial hypertension (PAH)/right ventricular dysfunction is investigated in this article , where the authors investigated vegetable consumptions and metal levels between PAH patients and controls.
Abstract: Abstract Pulmonary arterial hypertension (PAH) prevalence is increasing worldwide, and the prognosis is poor with 5‐year survival < 50% in high risk patients. The relationship between metal exposure/essential metal dyshomeostasis and PAH/right ventricular dysfunction is less investigated. The aim of this study is to investigate vegetable consumptions and metal levels between PAH patients and controls. This was a prospective, single center pilot study. Questionnaires were completed by all study subjects (20 PAH patients and 10 healthy controls) on smoking, metal exposure risks, metal supplements, and vegetable consumptions. Blood and urine samples were collected to measure 25 metal levels in blood, plasma, and urine using an X Series II quadrupole inductively coupled plasma mass spectrometry. Statistical analysis was conducted using SAS 9.5 and results with p value < 0.05 were considered significant. Vegetables consumptions (broccoli risk ratio [RR] = 0.4, CI = (0.2, 0.9)], cabbage [RR = 0.2, CI = (0.1, 0.8)], and brussel sprouts [RR = 0.2, CI = (0.1, 0.5)]) are associated with less risks of PAH. In the plasma samples, silver (p < 0.001), and copper (p = 0.002) levels were significantly higher in PAH patients. There was significant positive correlation between cardiac output and cardiac index with plasma levels of silver (r = 0.665, p = 0.001 and r = 0.678 p = 0.001), respectively. There was significant correlation between mixed venous saturation, 6‐min walk distance, and last BNP with plasma levels of chromium (r = −0.520, p = 0.022; r = −0.55, p = 0.014; r = 0.463, p = 0.039), respectively. In conclusion, there are significant differences between PAH and control groups in terms of vegetable consumptions and metal concentrations. Silver and chromium levels are correlated with clinical indicators of PAH severities.
TL;DR: Li et al. as mentioned in this paper proposed a leakage detection and localization approach by integrating the attention mechanism (AM) with the LSTM network, which assigns a higher attention weight to the sensor close to the leakage position, indicating the variation of data from the sensor has a significant influence on leakage localization.
Abstract: Long short-term memory (LSTM) has been widely applied to real-time automated natural gas leak detection and localization. However, LSTM approach could not provide the interpretation that this leak position is localized instead of other positions. This study proposes a leakage detection and localization approach by integrating the attention mechanism (AM) with the LSTM network. In this hybrid network, a fully-connected neural network behaving as AM is first applied to assign initial weights to time-series data. LSTM is then used to discover the complex correlation between the weighted data and leakage positions. A labor-scale pipeline leakage experiment of an urban natural gas distribution network is conducted to construct the benchmark dataset. A comparison between the proposed approach and the state-of-the-arts is also performed. The results demonstrate our proposed approach exhibits higher accuracy with AUC = 0.99. Our proposed approach assigns a higher attention weight to the sensor close to the leakage position, indicating the variation of data from the sensor has a significant influence on leakage localization. It corresponds that the closer to the leakage position, the larger variation of monitoring pressure after leakage, which enhances the detection results’ trustiness. This study provides a transparent and robust alternative for real-time automatic pipeline leak detection and localization, which contributes to constructing a digital twin of emergency management of urban pipeline leakage.
TL;DR: Wang et al. as discussed by the authors proposed a structure-aware registration method by incorporating structural information of related organs with segmentation-guided deep registration network, which extracts structural information in hierarchical geometric perspectives of line and surface.
Abstract: Image registration of liver dynamic contrast-enhanced computed tomography (DCE-CT) is crucial for diagnosis and image-guided surgical planning of liver cancer. However, intensity variations due to the flow of contrast agents combined with complex spatial motion induced by respiration brings great challenge to existing intensity-based registration methods. To address these problems, we propose a novel structure-aware registration method by incorporating structural information of related organs with segmentation-guided deep registration network. Existing segmentation-guided registration methods only focus on volumetric registration inside the paired organ segmentations, ignoring the inherent attributes of their anatomical structures. In addition, such paired organ segmentations are not always available in DCE-CT images due to the flow of contrast agents. Different from existing segmentation-guided registration methods, our proposed method extracts structural information in hierarchical geometric perspectives of line and surface. Then, according to the extracted structural information, structure-aware constraints are constructed and imposed on the forward and backward deformation field simultaneously. In this way, all available organ segmentations, including unpaired ones, can be fully utilized to avoid the side effect of contrast agent and preserve the topology of organs during registration. Extensive experiments on an in-house liver DCE-CT dataset and a public LiTS dataset show that our proposed method can achieve higher registration accuracy and preserve anatomical structure more effectively than state-of-the-art methods.