Search

Transformation towards a digital government imposes significant demands on the capabilities of legacy infrastructure. We closely followed a Swedish municipality that designed and implemented a solution to improve the building permit application process with an aim to improve citizen service. We developed six design principles (DPs): availability, timeliness, actionability, transparency, personalization, and generalizability. These DPs guide the solution design and provide a seamless application experience for citizens and business owners. We also discuss the reasoning behind the design choices and the implications of the solution. The artifact encompasses understanding citizens’ needs, identifying constraints of the legacy systems, formulating design principles, and developing architectural designs. However, addressing the social aspects of legacy systems, such as organizational culture change, necessitates additional steps, and is worth investigation in future studies.

This paper aims to explore how to build a sustainable peace and development model for China’s peacekeeping efforts through the application of data-driven methods from UN Global Pulse. UN Global Pulse is a United Nations agency dedicated to using big data and artificial intelligence technologies to address global challenges. In this paper, we will introduce the working principles of UN Global Pulse and its application in the fields of peacekeeping and development. Then, we will discuss the current situation of China’s participation in peacekeeping operations and how data-driven methods can help China play a greater role in peacekeeping tasks. Finally, we will propose a sustainable peace and development model that combines data-driven methods with the advantages of China’s peacekeeping efforts to achieve long-term peace and development goals.

Background and Objectives Structural imaging can offer insights into the cortical morphometry of migraine, which might reflect adaptations to recurring nociceptive messaging. This study compares cortical morphometry between a large sample of people with migraine and healthy controls, as well as across migraine subtypes. Methods Adult participants with migraine and age-matched and sex-matched healthy controls attended a single MRI session with magnetization-prepared rapid acquisition gradient echo and fluid-attenuated inversion recovery sequences at 3T. Cortical surface area, thickness, and volume were compared between participants with migraine (including subgroups) and healthy controls across the whole cortex within FreeSurfer and reported according to the Desikan-Killiany atlas. The analysis used cluster-determining thresholds of p < 0.0001 and cluster-wise thresholds of p < 0.05, adjusted for age, sex, and total intracranial volume. Results A total of 296 participants with migraine (mean age 41.6 years ± 12.4 SD, 261 women) and 155 healthy controls (mean age 41.1 years ± 11.7 SD, 133 women) were included. Among the participants with migraine, 180 (63.5%) had chronic migraine, 103 (34.8%) had migraine with aura, and 88 (29.7%) experienced a migraine headache during the scan. The total cohort of participants with migraine had reduced cortical surface area in the left insula, compared with controls (p < 0.0001). Furthermore, participants with chronic migraine (n = 180) exhibited reduced surface area in the left insula (p < 0.0001) and increased surface area in the right caudal anterior cingulate cortex (p < 0.0001), compared with controls. We found no differences specific to participants with aura or ongoing migraine headache. Post hoc tests revealed a positive correlation between monthly headache days and surface area within the identified anterior cingulate cluster (p = 0.014). Discussion The identified cortical changes i, BACKGROUND AND OBJECTIVES: Structural imaging can offer insights into the cortical morphometry of migraine, which might reflect adaptations to recurring nociceptive messaging. This study compares cortical morphometry between a large sample of people with migraine and healthy controls, as well as across migraine subtypes. METHODS: Adult participants with migraine and age-matched and sex-matched healthy controls attended a single MRI session with magnetization-prepared rapid acquisition gradient echo and fluid-attenuated inversion recovery sequences at 3T. Cortical surface area, thickness, and volume were compared between participants with migraine (including subgroups) and healthy controls across the whole cortex within FreeSurfer and reported according to the Desikan-Killiany atlas. The analysis used cluster-determining thresholds of p < 0.0001 and cluster-wise thresholds of p < 0.05, adjusted for age, sex, and total intracranial volume. RESULTS: A total of 296 participants with migraine (mean age 41.6 years ± 12.4 SD, 261 women) and 155 healthy controls (mean age 41.1 years ± 11.7 SD, 133 women) were included. Among the participants with migraine, 180 (63.5%) had chronic migraine, 103 (34.8%) had migraine with aura, and 88 (29.7%) experienced a migraine headache during the scan. The total cohort of participants with migraine had reduced cortical surface area in the left insula, compared with controls (p < 0.0001). Furthermore, participants with chronic migraine (n = 180) exhibited reduced surface area in the left insula (p < 0.0001) and increased surface area in the right caudal anterior cingulate cortex (p < 0.0001), compared with controls. We found no differences specific to participants with aura or ongoing migraine headache. Post hoc tests revealed a positive correlation between monthly headache days and surface area within the identified anterior cingulate cluster (p = 0.014). DISCUSSION: The identified cortical changes in migraine were limit

The split-and-pool method has been widely used to synthesize chemical libraries of a large size for early drug discovery, albeit without the possibility of meaningful quality control. In contrast, a self-assembled DNA-encoded chemical library (DEL) allows us to construct an m x n-member library by mixing an m-member and an n-member pre-purified sub-library. Herein, we report a trio-pharmacophore DEL (T-DEL) of m x l x n members through assembling three pre-purified and validated sub-libraries. The middle sub-library is synthesized using DNA-templated synthesis with different reaction mechanisms and designed as a linkage connecting the fragments displayed on the flanking two sub-libraries. Despite assembling three fragments, the resulting compounds do not exceed the up-to-date standard of molecular weight regarding drug-likeness. We demonstrate the utility of T-DEL in linker optimization for known binding fragments against trypsin and carbonic anhydrase II and by de novo selections against matrix metalloprotease-2 and −9.

type:Article, A newly developed factor analysis, origin-shifted factor analysis, was compared with a normal factor analysis to analyze the spectral changes of English speech. Our first aim was to investigate whether these analyses would cause differences in the factor loadings and the extracted spectral-factor scores. The methods mainly differed in whether to use cepstral liftering and an origin shift. The results showed that three spectral factors were obtained in four main frequency bands, but neither the cepstral liftering nor the origin shift distorted the essential characteristics of the factors. This confirms that the origin-shifted factor analysis is more recommendable for future speech analyses, since it would reduce the generation of noise in resynthesized speech. Our second aim was to further identify acoustic correlates of English phonemes. Our data show for the first time that the distribution of obstruents in English speech constitutes an L-shape related to two spectral factors on the three-dimensional configuration. One factor had center loadings around 4100 Hz, while the other was bimodal with peaks around 300 Hz and 2300 Hz. This new finding validates the use of multivariate analyses to connect English phonology and speech acoustics.

Shape learning, or the ability to leverage shape information, could be a desirable property of convolutional neural networks (CNNs) when target objects have specific shapes. While some research on the topic is emerging, there is no systematic study to conclusively determine whether and under what circumstances CNNs learn shape. Here, we present such a study in the context of segmentation networks where shapes are particularly important. We define shape and propose a new behavioral metric to measure the extent to which a CNN utilizes shape information. We then execute a set of experiments with synthetic and real-world data to progressively uncover under which circumstances CNNs learn shape and what can be done to encourage such behavior. We conclude that (i) CNNs do not learn shape in typical settings but rather rely on other features available to identify the objects of interest, (ii) CNNs can learn shape, but only if the shape is the only feature available to identify the object, (iii) sufficiently large receptive field size relative to the size of target objects is necessary for shape learning; (iv) a limited set of augmentations can encourage shape learning; (v) learning shape is indeed useful in the presence of out-of-distribution data., Comment: Accepted by Pattern Recognition, reformatted for Arxiv for readability

Training segmentation models for medical images continues to be challenging due to the limited availability of data annotations. Segment Anything Model (SAM) is a foundation model that is intended to segment user-defined objects of interest in an interactive manner. While the performance on natural images is impressive, medical image domains pose their own set of challenges. Here, we perform an extensive evaluation of SAM's ability to segment medical images on a collection of 19 medical imaging datasets from various modalities and anatomies. We report the following findings: (1) SAM's performance based on single prompts highly varies depending on the dataset and the task, from IoU=0.1135 for spine MRI to IoU=0.8650 for hip X-ray. (2) Segmentation performance appears to be better for well-circumscribed objects with prompts with less ambiguity and poorer in various other scenarios such as the segmentation of brain tumors. (3) SAM performs notably better with box prompts than with point prompts. (4) SAM outperforms similar methods RITM, SimpleClick, and FocalClick in almost all single-point prompt settings. (5) When multiple-point prompts are provided iteratively, SAM's performance generally improves only slightly while other methods' performance improves to the level that surpasses SAM's point-based performance. We also provide several illustrations for SAM's performance on all tested datasets, iterative segmentation, and SAM's behavior given prompt ambiguity. We conclude that SAM shows impressive zero-shot segmentation performance for certain medical imaging datasets, but moderate to poor performance for others. SAM has the potential to make a significant impact in automated medical image segmentation in medical imaging, but appropriate care needs to be applied when using it., Comment: Link to our code: https://github.com/mazurowski-lab/segment-anything-medical-evaluation

The continuous improvement of human-computer interaction technology makes it possible to compute emotions. In this paper, we introduce our submission to the CVPR 2023 Competition on Affective Behavior Analysis in-the-wild (ABAW). Sentiment analysis in human-computer interaction should, as far as possible Start with multiple dimensions, fill in the single imperfect emotion channel, and finally determine the emotion tendency by fitting multiple results. Therefore, We exploited multimodal features extracted from video of different lengths from the competition dataset, including audio, pose and images. Well-informed emotion representations drive us to propose a Attention-based multimodal framework for emotion estimation. Our system achieves the performance of 0.361 on the validation dataset. The code is available at [https://github.com/xkwangcn/ABAW-5th-RT-IAI]., Comment: 5 pages, 1 figures

Weakly-supervised temporal action localization aims to localize action instances in videos with only video-level action labels. Existing methods mainly embrace a localization-by-classification pipeline that optimizes the snippet-level prediction with a video classification loss. However, this formulation suffers from the discrepancy between classification and detection, resulting in inaccurate separation of foreground and background (F\&B) snippets. To alleviate this problem, we propose to explore the underlying structure among the snippets by resorting to unsupervised snippet clustering, rather than heavily relying on the video classification loss. Specifically, we propose a novel clustering-based F\&B separation algorithm. It comprises two core components: a snippet clustering component that groups the snippets into multiple latent clusters and a cluster classification component that further classifies the cluster as foreground or background. As there are no ground-truth labels to train these two components, we introduce a unified self-labeling mechanism based on optimal transport to produce high-quality pseudo-labels that match several plausible prior distributions. This ensures that the cluster assignments of the snippets can be accurately associated with their F\&B labels, thereby boosting the F\&B separation. We evaluate our method on three benchmarks: THUMOS14, ActivityNet v1.2 and v1.3. Our method achieves promising performance on all three benchmarks while being significantly more lightweight than previous methods. Code is available at https://github.com/Qinying-Liu/CASE, Comment: ICCV2023

Deep neural networks (DNNs) have been deployed for many image segmentation tasks and achieved outstanding performance. However, preparing a dataset for training segmentation DNNs is laborious and costly since typically pixel-level annotations are provided for each object of interest. To alleviate this issue, one can provide only weak labels such as bounding boxes or scribbles, or less accurate (noisy) annotations of the objects. These are significantly faster to generate and thus result in more annotated images given the same time budget. However, the reduction in quality might negatively affect the segmentation performance of the resulting model. In this study, we perform a thorough cost-effectiveness evaluation of several weak and noisy labels. We considered 11 variants of annotation strategies and 4 datasets. We conclude that the common practice of accurately outlining the objects of interest is virtually never the optimal approach when the annotation time is limited, even if notable annotation time is available (10s of hours). Annotation approaches that stood out in such scenarios were (1) contour-based annotation with rough continuous traces, (2) polygon-based annotation with few vertices, and (3) box annotations combined with the Segment Anything Model (SAM). In situations where unlimited annotation time was available, precise annotations still lead to the highest segmentation model performance.

Computer simulations offer a robust toolset for exploring complex systems across various disciplines. A particularly impactful approach within this realm is Agent-Based Modeling (ABM), which harnesses the interactions of individual agents to emulate intricate system dynamics. ABM's strength lies in its bottom-up methodology, illuminating emergent phenomena by modeling the behaviors of individual components of a system. Yet, ABM has its own set of challenges, notably its struggle with modeling natural language instructions and common sense in mathematical equations or rules. This paper seeks to transcend these boundaries by integrating Large Language Models (LLMs) like GPT into ABM. This amalgamation gives birth to a novel framework, Smart Agent-Based Modeling (SABM). Building upon the concept of smart agents -- entities characterized by their intelligence, adaptability, and computation ability -- we explore in the direction of utilizing LLM-powered agents to simulate real-world scenarios with increased nuance and realism. In this comprehensive exploration, we elucidate the state of the art of ABM, introduce SABM's potential and methodology, and present three case studies (source codes available at https://github.com/Roihn/SABM), demonstrating the SABM methodology and validating its effectiveness in modeling real-world systems. Furthermore, we cast a vision towards several aspects of the future of SABM, anticipating a broader horizon for its applications. Through this endeavor, we aspire to redefine the boundaries of computer simulations, enabling a more profound understanding of complex systems., Comment: Source codes are available at https://github.com/Roihn/SABM

In this article, we study the collocated and distributed deployment of intelligent reflecting surfaces (IRS) for a fixed total number of IRS elements to support enhanced mobile broadband (eMBB) and ultra-reliable low-latency communication (URLLC) services inside a factory. We build a channel model that incorporates the line-of-sight (LOS) probability and power loss of each transmission path, and propose three metrics, namely, the expected received signal-to-noise ratio (SNR), expected finite-blocklength (FB) capacity, and expected outage probability, where the expectation is taken over the probability distributions of interior blockages and channel fading. The expected received SNR and expected FB capacity for extremely high blockage densities are derived in closed-form as functions of the amount and height of IRSs and the density, size, and penetration loss of blockages, which are verified by Monte Carlo simulations. Results show that deploying IRSs vertically higher leads to higher expected received SNR and expected FB capacity. By analysing the average/minimum/maximum of the three metrics versus the number of IRSs, we find that for high blockage densities, both eMBB and URLLC services benefit from distributed deployment; and for low blockage densities, URLLC services benefit from distributed deployment while eMBB services see limited difference between collocated and distributed deployment.

Instances of casualties resulting from large crowds persist, highlighting the existing limitations of current crowd management practices in Smart Cities. One notable drawback is the insufficient provision for disadvantaged individuals who may require additional time to evacuate due to their slower running speed. Moreover, the existing escape strategies may fall short of ensuring the safety of all individuals during a crowd surge. To address these pressing concerns, this paper proposes two crowd management methodologies. Firstly, we advocate for implementing a fair evacuation strategy following a surge event, which considers the diverse needs of all individuals, ensuring inclusivity and mitigating potential risks. Secondly, we propose a preventative approach involving the adjustment of attraction locations and switching between stage performances in large-crowded events to minimize the occurrence of surges and enhance crowd dispersion. We used high-fidelity crowd management simulators to assess the effectiveness of our proposals. Our findings demonstrate the positive impact of the fair evacuation strategy on safety measures and inclusivity, which increases fairness by 41.8% on average. Furthermore, adjusting attraction locations and stage performances has shown a significant reduction in surges by 34% on average, enhancing overall crowd safety.

Source-free object detection (SFOD) aims to adapt a source-trained detector to an unlabeled target domain without access to the labeled source data. Current SFOD methods utilize a threshold-based pseudo-label approach in the adaptation phase, which is typically limited to high-confidence pseudo-labels and results in a loss of information. To address this issue, we propose a new approach to take full advantage of pseudo-labels by introducing high and low confidence thresholds. Specifically, the pseudo-labels with confidence scores above the high threshold are used conventionally, while those between the low and high thresholds are exploited using the Low-confidence Pseudo-labels Utilization (LPU) module. The LPU module consists of Proposal Soft Training (PST) and Local Spatial Contrastive Learning (LSCL). PST generates soft labels of proposals for soft training, which can mitigate the label mismatch problem. LSCL exploits the local spatial relationship of proposals to improve the model's ability to differentiate between spatially adjacent proposals, thereby optimizing representational features further. Combining the two components overcomes the challenges faced by traditional methods in utilizing low-confidence pseudo-labels. Extensive experiments on five cross-domain object detection benchmarks demonstrate that our proposed method outperforms the previous SFOD methods, achieving state-of-the-art performance.

Stem cell bioengineering and therapy require different model systems and materials in different stages of development. If a chemically defined biomatrix system can fulfill most tasks, it can minimize the discrepancy among various setups. By screening biomaterials synthesized through a coacervation-mediated self-assembling mechanism, a biomatrix system optimal for 2D human mesenchymal stromal cell (hMSC) culture and osteogenesis is identified. Its utility for hMSC bioengineering is further demonstrated in coating porous bioactive glass scaffolds and nanoparticle synthesis for esiRNA delivery to knock down the SOX-9 gene with high delivery efficiency. The self-assembled injectable system is further utilized for 3D cell culture, segregated co-culture of hMSC with human umbilical vein endothelial cells (HUVEC) as an angiogenesis model, and 3D bioprinting. Most interestingly, the coating of bioactive glass with the self-assembled biomatrix not only supports the proliferation and osteogenesis of hMSC in the 3D scaffold but also induces the amorphous bioactive glass (BG) scaffold surface to form new apatite crystals resembling bone-shaped plate structures. Thus, the self-assembled biomatrix system can be utilized in various dimensions, scales, and geometries for many different bioengineering applications.

Information processing functions are essential for organisms to perceive and react to their complex environment, and for humans to analyze and rationalize them. While our brain is extraordinary at processing complex information, winner-take-all, as a type of biased competition is one of the simplest models of lateral inhibition and competition among biological neurons. It has been implemented as DNAbased neural networks, for example, to mimic pattern recognition. However, the utility of DNA-based computation in information processing for real biotechnological applications remains to be demonstrated. In this paper, a biased competitionmethod for nonlinear manipulation and analysis ofmixtures of DNA sequences was developed. Unlike conventional biological experiments, selected species were not directly subjected to analysis. Instead, parallel computation among a myriad of different DNA sequences was carried out to reduce the information entropy. The method could be used for various oligonucleotideencoded libraries, as we have demonstrated its application in decoding and data analysis for selection experiments with DNA-encoded chemical libraries against protein targets.

Premature cardiac myocytes derived from human induced pluripotent stem cells (hiPSC-CMs) show heterogeneous action potentials (APs), probably due to different expression patterns of membrane ionic currents. We developed a method for determining expression patterns of functional channels in terms of whole-cell ionic conductance (Gx) using individual spontaneous AP configurations. It has been suggested that apparently identical AP configurations can be obtained using different sets of ionic currents in mathematical models of cardiac membrane excitation. If so, the inverse problem of Gx estimation might not be solved. We computationally tested the feasibility of the gradient-based optimization method. For a realistic examination, conventional 'cell-specific models' were prepared by superimposing the model output of AP on each experimental AP recorded by conventional manual adjustment of Gxs of the baseline model. Gxs of 4–6 major ionic currents of the 'cell-specific models' were randomized within a range of ± 5–15% and used as an initial parameter set for the gradient-based automatic Gxs recovery by decreasing the mean square error (MSE) between the target and model output. Plotting all data points of the MSE–Gx relationship during optimization revealed progressive convergence of the randomized population of Gxs to the original value of the cell-specific model with decreasing MSE. The absence of any other local minimum in the global search space was confirmed by mapping the MSE by randomizing Gxs over a range of 0.1–10 times the control. No additional local minimum MSE was obvious in the whole parameter space, in addition to the global minimum of MSE at the default model parameter.

The functionalization of the outer lattice of in vitro reconstituted microtubules has paved the way for the development of diverse nano-device applications. The outer lattice has been metallized for the bottom-up synthesis of nanowires composed of various materials. Moreover, a wide range of biomolecules and nanoprobes have been attached to the outer surface for nano-scale transport and detection assays in conjunction with motor proteins. The functionalization of the outer lattice has certain implications: While the nanowires adopt the overall shape of the microtubules, their surface is inhomogeneous due to the absence of any morphological control. The attachment of cargo on the outer lattice creates a ‘roadblock effect’ hindering the transport activity of the motor proteins as they share a common substrate surface. In this project, the utilization of the hollow interior region of the microtubules, called the lumen (∼15 nm in diameter) is proposed to overcome these limitations. A strategy is developed to functionalize the microtubule lumen by targeting molecular cargo conjugated to lumen-binding (anti-acetyl alpha-tubulin lysine-40) antibodies. This would optimize existing motility-based applications as the outer surface would be exclusively available for the activity of the motor proteins. Furthermore, microtubules functionalized with luminal gold nanoparticle ‘seeds’ are utilized for the lumen-templated assembly of gold nanowires.

The inward diffusion of metal electrodes is one of the main reasons for the deterioration of the long-term device stability of perovskite solar cells (PSCs). Thus, herein, we adopt a simple additive engineering strategy to modify the barrier material bathocuproine (BCP) with 1,3,5-triazine-2,4,6-trithiol trisodium salt (TTTS). Different from the traditional physical blocking strategies, TTTS could prevent the metal electrodes (e.g., gold, silver, and copper) from diffusing inward through the strong chemical coordination between TTTS and the metal electrode. The TTTS additive also improved the conductivity and band structure of BCP, thus enhancing the ability of BCP to extract electrons from the perovskite layer to the electrode. Consequently, the inverted device modified with TTTS exhibited a high efficiency of 22.59%, which is among the highest efficiencies reported to date for inverted PSCs. More importantly, it showed excellent operational, ambient, and thermal stability. The target device maintained its initial efficiency with no loss under continuous one-sun illumination at maximum power point tracking after 1000 h (the champion device), 91% in air (50% +/- 5% RH) for 5000 h, and 93% after heating at 85 degrees C for 1500 h (average efficiency from ten devices).

Additive engineering is an effective strategy for defect passivation and performance improvement of perovskite solar cells (PSCs). However, few additives have achieved outstanding stability with high efficiency by simultaneously passivating deep and shallow defects. Herein, we design a novel ionic silicone polymer (PECL) with multi-active sites as an additive to modify inverted PSCs. The C-O groups in the PECL polymer can chelate with undercoordinated Pb2+ and Pb clusters to passivate deep defects; and the ionic groups in the PECL polymer can generate electrostatic interaction with both positively and negatively charged vacancies, which help to repair shallow defects. Moreover, we quantitatively reveal the effect of deep and shallow defects on the efficiency and stability of PSCs separately, by establishing the correlation between additives with different functional groups and the performance of devices. Consequently, the power conversion efficiency of the PECL-modified inverted PSC increases from 20.02% to 23.11%. More importantly, the encapsulated PSCs maintain 95% of their initial steady-state power output after 1500 hours under AM 1.5 illumination at the maximum power point at 45 degrees C. Therefore, we provide a universal guideline of polymer structure design for defect healing in stabilizing PSCs with high efficiency.

Nickel oxide (NiOx) is a promising hole transport material in inverted organic-inorganic metal halide perovskite solar cells. However, its low intrinsic conductivity hinders its further improvement in device performance. Here, we employ a trimercapto-s-triazine trisodium salt (TTTS) as a chelating agent of Ni2+ in the NiOx layer to improve its conductivity. Due to the electron-deficient triazine ring, the TTTS complexes with Ni2+ in NiOx via a strong Ni2+-N coordination bond and increases the ratio of Ni3+:Ni2+. The increased Ni3+ concentration adjusts the band structure of NiOx, thus enhancing hole density and mobility, eventually improving the intrinsic conductivity of NiOx. As a result, the device with TTTS modification displays a champion power conversion efficiency (PCE) of 22.81%. The encapsulated device based on a modified-NiOx layer maintains 94% of its initial power output at the maximum power point and continuous one-sun illumination for 1000 h at 45 degrees C. In addition, the unencapsulated target devices also maintain 92% at 60 +/- 5% relative humidity and 25 degrees C in the air for 5000 h; and 91% at 85 degrees C in a nitrogen atmosphere for 1000 h. The research provides an effective strategy to enhance PCE and stability of inverted PSCs via modifying NiOx films with triazine molecule.

Because of the compatibility with tandem devices and the ability to be manufactured at low temperatures, inverted perovskite solar cells have generated far-ranging interest for potential commercial applications. However, their efficiency remains inadequate owing to various traps in the perovskite film and the restricted hole blocking ability of the electron transport layer. Thus, in this work, a wide-bandgap n-type semiconductor, 4,6-bis(3,5-di(pyridin-4-yl)phenyl)-2-phenylpyrimidine (B4PyPPM), to modify a perovskite film via an anti-solvent method is introduced. The nitrogen sites of pyrimidine and pyridine rings in B4PyPPM exhibit strong interactions with the undercoordinated lead ions in the perovskite material. These interactions can reduce the trap state densities and inhibit nonradiative recombination of the perovskite bulk. Moreover, B4PyPPM can partially aggregate on the perovskite surface, leading to an improvement in the hole-blocking ability at its interface. This modification can also increase the built-in potential and upshift the Fermi level of the modified perovskite film, promoting electron extraction to the electron transport layer. The champion device achieves a high efficiency of 23.51%. Meantime, the sealed device retains approximate to 80% of its initial performance under a maximum power point tracking for nearly 2400 h, demonstrating an excellent operational stability.

Dapeng Zhou,1,2,&ast; Guangtao Huang,1– 4,&ast; Guangchao Xu,1,2 Lijuan Xiang,5 Siyi Huang,1,2 Xinchong Chen,1,2 Yixin Zhang,4 Dali Wang1,2 1Department of Burns and Plastic Surgery, Affiliated Hospital of Zunyi Medical University, Zunyi, People’s Republic of China; 2The Collaborative Innovation Center of Tissue Damage Repair and Regeneration Medicine of Zunyi Medical University, Zunyi, People’s Republic of China; 3Department of Burn and Plastic Surgery, Department of Wound Repair, Shenzhen Institute of Translational Medicine, The First Affiliated Hospital of Shenzhen University, Shenzhen Second People’s Hospital, Shenzhen, People’s Republic of China; 4Department of Plastic and Reconstructive Surgery, Shanghai Ninth People’s Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, People’s Republic of China; 5Department of Clinical Laboratory, Shenzhen Institute of Translational Medicine, The First Affiliated Hospital of Shenzhen University, Shenzhen Second People’s Hospital, Shenzhen, People’s Republic of China&ast;These authors contributed equally to this workCorrespondence: Yixin Zhang, Department of Plastic and Reconstructive Surgery, Shanghai Ninth People’s Hospital, Shanghai Jiao Tong University School of Medicine, 639 Zhizaoju Road, Shanghai, 200011, People’s Republic of China, Tel +86-21-23271699, Email zhsh06089@163.com Dali Wang, Department of Burns and Plastic Surgery, Affiliated Hospital of Zunyi Medical University, 149 Dalian Road, Huichuan District, Zunyi, 563003, People’s Republic of China, Tel +86-851-28645801, Email daliwangzy@sina.comPurpose: Pseudomonas aeruginosa is a common pathogen of infection in burn and trauma patients, and multi-drug resistant P. aeruginosa has become an increasingly important pathogen. Essential genes are key to the development of novel antibiotics. The PA0715 gene is a novel unidentified essential gene that has attracted our interest as a potential antibiotic target. Our stu

Microbes play a critical role in plant litter decomposition and influence the fate of carbon in rivers and riparian zones. When decomposing low-nutrient plant litter, microbes acquire nitrogen (N) and phosphorus (P) from the environment (i.e., nutrient immobilization), and this process is potentially sensitive to nutrient loading and changing climate. Nonetheless, environmental controls on immobilization are poorly understood because rates are also influenced by plant litter chemistry, which is coupled to the same environmental factors. Here we used a standardized, low-nutrient organic matter substrate (cotton strips) to quantify nutrient immobilization at 100 paired stream and riparian sites representing 11 biomes worldwide. Immobilization rates varied by three orders of magnitude, were greater in rivers than riparian zones, and were strongly correlated to decomposition rates. In rivers, P immobilization rates were controlled by surface water phosphate concentrations, but N immobilization rates were not related to inorganic N. The N:P of immobilized nutrients was tightly constrained to a molar ratio of 10:1 despite wide variation in surface water N:P. Immobilization rates were temperature-dependent in riparian zones but not related to temperature in rivers. However, in rivers nutrient supply ultimately controlled whether microbes could achieve the maximum expected decomposition rate at a given temperature. Collectively, we demonstrated that exogenous nutrient supply and immobilization are critical control points for decomposition of organic matter.

Nowadays, the epidemic (COVID-19) accelerates media convergence of films and streaming media. Under the background, this study mainly explores how algorithm technology affects the film industry. The objective of the study is to discuss the interventions that algorithmic technology produces in the traditional film ecosystem, using the film production process as a reference timeline to consider how creator-investors should make trade-offs between the rational stance of algorithms and the emotional creation of film art.

Citizen science and machine learning should be considered for monitoring the coastal and ocean environment due to the scale of threats posed by climate change and the limited resources to fill knowledge gaps. Using data from the annotation activity of citizen scientists in a Swedish marine project, we constructed Deep Neural Network models to predict forthcoming engagement. We tested the models to identify patterns in annotation engagement. Based on the results, it is possible to predict whether an annotator will remain active in future sessions. Depending on the goals of individual citizen science projects, it may also be necessary to identify either those volunteers who will leave or those who will continue annotating. This can be predicted by varying the threshold for the prediction. The engagement metrics used to construct the models are based on time and activity and can be used to infer latent characteristics of volunteers and predict their task interest based on their activity patterns. They can estimate if volunteers can accomplish a given number of tasks in a certain amount of time, identify early on who is likely to become a top contributor or identify who is likely to quit and provide them with targeted interventions. The novelty of our predictive models lies in the use of Deep Neural Networks and the sequence of volunteer annotations. A limitation of our models is that they do not use embeddings constructed from user profiles as input data, as many recommender systems do. We expect that including user profiles would improve prediction performance., Comment: Working paper, 8 pages

Domain adaptation (DA) aims to transfer knowledge from a label-rich source domain to a related but label-scarce target domain. The conventional DA strategy is to align the feature distributions of the two domains. Recently, increasing researches have focused on self-training or other semi-supervised algorithms to explore the data structure of the target domain. However, the bulk of them depend largely on confident samples in order to build reliable pseudo labels, prototypes or cluster centers. Representing the target data structure in such a way would overlook the huge low-confidence samples, resulting in sub-optimal transferability that is biased towards the samples similar to the source domain. To overcome this issue, we propose a novel contrastive learning method by processing low-confidence samples, which encourages the model to make use of the target data structure through the instance discrimination process. To be specific, we create positive and negative pairs only using low-confidence samples, and then re-represent the original features with the classifier weights rather than directly utilizing them, which can better encode the task-specific semantic information. Furthermore, we combine cross-domain mixup to augment the proposed contrastive loss. Consequently, the domain gap can be well bridged through contrastive learning of intermediate representations across domains. We evaluate the proposed method in both unsupervised and semi-supervised DA settings, and extensive experimental results on benchmarks reveal that our method is effective and achieves state-of-the-art performance. The code can be found in https://github.com/zhyx12/MixLRCo., Comment: 11 pages, 5 figures; an outdated version that will be updated

The sequential recommendation systems capture users' dynamic behavior patterns to predict their next interaction behaviors. Most existing sequential recommendation methods only exploit the local context information of an individual interaction sequence and learn model parameters solely based on the item prediction loss. Thus, they usually fail to learn appropriate sequence representations. This paper proposes a novel recommendation framework, namely Graph Contrastive Learning for Sequential Recommendation (GCL4SR). Specifically, GCL4SR employs a Weighted Item Transition Graph (WITG), built based on interaction sequences of all users, to provide global context information for each interaction and weaken the noise information in the sequence data. Moreover, GCL4SR uses subgraphs of WITG to augment the representation of each interaction sequence. Two auxiliary learning objectives have also been proposed to maximize the consistency between augmented representations induced by the same interaction sequence on WITG, and minimize the difference between the representations augmented by the global context on WITG and the local representation of the original sequence. Extensive experiments on real-world datasets demonstrate that GCL4SR consistently outperforms state-of-the-art sequential recommendation methods., Comment: 8 pages, 3 figures, Accepted by IJCAI 2022

Muqing Zhang,1,2,&ast; Jian Chen,3,4,&ast; Yanwei Wang,2 Guobin Kang,2 Yixin Zhang,3,4 Xue Han3,4 1College of Integrative Medicine, Hebei University of Chinese Medicine, Shijiazhuang, People’s Republic of China; 2Affiliated Hospital, Hebei University of Chinese Medicine, Shijiazhuang, People’s Republic of China; 3School of Pharmacy, Hebei University of Chinese Medicine, Shijiazhuang, People’s Republic of China; 4International Joint Research Center on Resource Utilization and Quality Evaluation of Traditional Chinese Medicine, Shijiazhuang, People’s Republic of China&ast;These authors contributed equally to this workCorrespondence: Yixin Zhang; Xue Han, Tel +86 311 89926316, Fax +86 311 89926316, Email hbzyx123@163.com; hanxuecc@126.comPurpose: The network pharmacology approach and validation experiment were performed to investigate the potential mechanisms of Agrimonia pilosa Ledeb. (APL) extract against acute myocardial infarction (AMI).Methods: The primary compounds of APL extract were identified by High-Performance Liquid Chromatography (HPLC) analysis. The intersecting targets of active compounds and AMI were determined via network pharmacology analysis. A mouse model of AMI was established by subcutaneous injection of isoproterenol (Iso). Mice were treated with APL extract by intragastric administration. We assessed the effects of APL extract on the electrocardiography (ECG), cardiac representative markers, representative indicators of oxidative stress, pathological changes, and phosphatidylinositol-3-kinase (PI3K)/protein kinase B (Akt) signaling pathway, as well as apoptosis-related indicators in the mice.Results: Five candidate compounds were identified in APL extract. Enrichment analyses indicated that APL extract could exert myocardial protective effects via the PI3K/Akt pathway. ST segment elevation and increased heart rate were obviously reversed in APL extract groups compared to Iso group. We also detected significant decreases in la

Qinlin Liu,1 Yixin Zhang,1 Chuan Hu,2 Dongli Yuan,3 Kuiyun Wang,4 Wen Fan,5 Feibao Pan,6 Qin Li,7 Yunfeng Wang,1 Ge Tan1 1Department of Neurology, The First Affiliated Hospital of Chongqing Medical University, Chongqing, People’s Republic of China; 2Department of Neurology, People’s Hospital of Bishan District, Bishan, People’s Republic of China; 3Department of Intelligent Medical Systems, Chongqing Medical University, Chongqing, People’s Republic of China; 4Department of Neurology, Jintang First People’s Hospital, Sichuan, People’s Republic of China; 5Department of Neurology, Guizhou Provincial People’s Hospital, Guiyang, People’s Republic of China; 6Department of Neurology, Suining Central Hospital, Suining, People’s Republic of China; 7Department of Neurology, Chongqing General Hospital, Chongqing, People’s Republic of ChinaCorrespondence: Ge Tan, The First Affiliated Hospital of Chongqing Medical University, 1 Youyi Road, Chongqing, People’s Republic of China, Email 273346692@qq.comPurpose: To explore changes of depression, anxiety and sleep disturbance in patients with episodic cluster headache inside and outside the attack period and assess the therapy to improve the treatment.Patients and Methods: We prospectively recruited 396 patients from 11 specialized headache outpatients and analyzed their headache characteristics, Hospital Anxiety Depression Scale scores, Pittsburgh sleep quality index scores, and the usage of psychiatric medications during as well as 1 month after the attack period.Results: A total of 220 patients completed the follow-up, 52.73% of whom had anxiety, 47.27% had depression and 49.09% had sleep disturbance inside the attack period. At follow-up, the percentage of these patients significantly decreased to 16.36%, 21.82% and 14.55% in the remission period, respectively (p < 0.05). Antidepressants and mood stabilizers were prescribed to 58.18% of the patients. However, both of the changes of Hospital Anxiety Depress

Yeletai Nurzati, Zhu Zhu, Heng Xu, Yixin Zhang Department of Plastic and Reconstructive Surgery, Shanghai Ninth People’s Hospital, Shanghai JiaoTong University School of Medicine, Shanghai, 201100, People’s Republic of ChinaCorrespondence: Heng Xu; Yixin Zhang Email xuh1990@gmail.com; zhangyx1576@163.comIntroduction: Keloid is a pathological scar type, which invades normal surrounding tissue without self-limiting to cause pain, itching, cosmetic disfigurement, etc. Knowledge of the molecular mechanisms underlying keloid remains unclear. This dilemma leads to no biomarker available for diagnosis. Thus, to seek accurate diagnosis, biomarkers are necessary for keloid diagnosis to help control its incidence.Methods: Gene Expression Omnibus (GEO) database was used to select differentially expressed miRNAs (DE-miRNAs) in GSE113620. miRTarBase miRNA–target tools were used to predict the interactions between miRNAs and their target mRNAs. Target mRNAs that were differentially expressed in keloid were selected by analyzing differentially expressed genes (DEGs) in GSE44270 and GSE92566. PPI network analysis, gene enrichment analysis, cell-specific and tissue-specific expression analyses of DE-target mRNAs were conducted. RT-PCR analysis was conducted to validate our results.Results: Three novel miRNAs (miR-30b-5p, miR-212-3p, miR-149-5p) and five target mRNAs (SIX1, CCNA2, CCNB1, FOXM1, RUNX2) were identified as potential biomarkers for keloid patients. Additionally, the potential functions of those miRNAs-mRNAs pathways were analyzed.Discussion: These findings of keloid-related miRNAs, mRNAs, and miRNA–mRNAs regulatory networks may provide insights into the underlying pathogenesis of keloid and serve as potential biomarkers for keloid diagnosis.Keywords: diagnostic biomarkers, miRNA–mRNA network, keloids, differentially expressed genes

Numerous studies on Western Opera singing have shown that listeners’ vowel identification performance decreases with an increasing fundamental frequency (fo). This study explores the intelligibility of high-pitched vowels in Yue Opera, the largest dialectal opera in China. Six long vowels (/i y e a o u/) were recorded by a professional female singer at ten fos between 220 and 932 Hz, of which 700-ms nuclei with flat fo contours and resonance trajectories were extracted as stimuli. In a within-subject design, sixteen phonetically trained listeners responded on a free-choice vowel quadrilateral (task 1) and in a two-alternative forced-choice task (task 2) to indicate which vowel was presented. Results show that vowels cluster in the perceptual space into three groups (/i y e/, /u o/, /a/) above 521 Hz and that listeners could identify vowels between but not within groups with high accuracy up to at least 932 Hz. Multidimensional scaling (MDS) of simulated auditory excitation patterns reveals highly differentiable spectral shapes between groups. These findings put into question whether previous results on Western Opera could be generalized to other forms of opera singing.

In this design, we propose a new method to process the large static signal in the received signal of Doppler ultrasound brain imaging. Through a reproduction-compensation method, a medium resolution ADC and a DAC are used to replace the high-resolution ADC used in the past for the dynamic Doppler signal observation.

Shabestar basin in the East Azerbaijan province, Northwest Iran, where irrigation is the main groundwater consumer, has experienced large-scale subsidence and groundwater deletion, which poses a threat to the local agricultural activities, economic development, and food security. With the emergency of mitigating the risk, satisfying future demand for groundwater, and improving resilience considering climate change, this study proposes a satellite-based approach to explore the spatio-temporal relationships between measured subsidence and hydrological variables in the basin to assist groundwater management strategy. We investigated ground subsidence in the basin using the SBAS-InSAR technique based on series of Sentinel-1A TOPS Synthetic Aperture Radar (SAR) images acquired from 22 January 2016 to 21 October 2020 along ascending and descending tracks. The study showed average subsidence rates ranged from -97.5 mm/year to 10 mm/year in the basin after decomposing line-of-sight velocity fields to vertical components. The prominent subsidence was found in the eastern and western portions of the basin, and the maximum average subsidence rate was detected in the eastern part of the basin near Nazarlu. Correlation analysis between the surface subsidence and potential driving factors, including the actual evapotranspiration (ETa), land surface temperature (LST), the normalized difference vegetation index (NDVI), precipitation (P), and soil water index (SWI), revealed a significant relationship between the first three variables and observed subsidence by InSAR. A multivariate long short-term memory (LSTM) network was established to investigate the importance of the first three variables and predict subsidence in the near future. The result quantitatively revealed that the agricultural practice had a major impact on subsidence occurrence in the basin. Furthermore, our findings indicated that the area is estimated to continue subsiding dramatically in the next five years. This, Groundwater is a critical natural resource. In the mid-latitude arid and semi-arid country, such as Iran lacking adequate surface water supply, groundwater has an important role in providing a water supply for irrigation and drinking water. The increasing groundwater demand in the face of population growth and climate change causes land subsidence, posing a threat to the local agricultural activities, economic development, and food security. In order to mitigate the risk and satisfy future water demand, the study proposes a satellite-based method to explore relationships between measured subsidence and hydrological variables, spatially and temporally, to assist groundwater management strategy. Shabestar basin, located in East Azerbaijan province, Northwest Iran, is selected as the study area. The basin has experienced large-scale subsidence and groundwater deletion in the past decades. We firstly investigated the local land subsidence from 2016 to 2020 using a mature technique called SBAS-InSAR technique. The result showed that mean subsidence rates ranged from -97.5 mm/year to 10 mm/year in the basin. The prominent subsidence was found in the eastern and western parts of the basin, and the largest mean subsidence rate was detected in the eastern part of the basin near Nazarlu town. Correlation analysis between the surface subsidence and five potential driving factors, including the actual evapotranspiration (ETa), land surface temperature (LST), the normalized difference vegetation index (NDVI), precipitation (P), and soil water index (SWI), indicated a significant relationship between the first three variables and measured subsidence. In order to assess the importance of variables and predict subsidence, in the near future, we constructed a deep learning model called a multivariate long short-term memory (LSTM) network. Our findings quantitatively revealed that the agricultural practice had a primary impact on subsidence occurrence in the basin. Furthermore, the r

Vertical sleeve gastrectomy (VSG) is one of the most effective and durable therapies for morbid obesity and its related complications. Although bile acids (BAs) have been implicated as downstream mediators of VSG, the specific mechanisms through which BA changes contribute to the metabolic effects of VSG remain poorly understood. Here, we confirm that high fat diet-fed global farnesoid X receptor (Fxr) knockout mice are resistant to the beneficial metabolic effects of VSG. However, the beneficial effects of VSG were retained in high fat diet-fed intestine- or liver-specific Fxr knockouts, and VSG did not result in Fxr activation in the liver or intestine of control mice. Instead, VSG decreased expression of positive hepatic Fxr target genes, including the bile salt export pump (Bsep) that delivers BAs to the biliary pathway. This reduced small intestine BA levels in mice, leading to lower intestinal fat absorption. These findings were verified in sterol 27-hydroxylase (Cyp27a1) knockout mice, which exhibited low intestinal BAs and fat absorption and did not show metabolic improvements following VSG. In addition, restoring small intestinal BA levels by dietary supplementation with taurocholic acid (TCA) partially blocked the beneficial effects of VSG. Altogether, these findings suggest that reductions in intestinal BAs and lipid absorption contribute to the metabolic benefits of VSG.

Cell membranes are essential to life. All living organisms adapt their membranes in response to environmental perturbations to maintain optimal membrane function. However, the principles underlying membrane lipid composition and membrane adaptation remain poorly understood. Deciphering the functional basis for the lipidome complexity employed by cellular life is central to the problem of understanding principles of cell membrane organization, and is currently limited by the staggering complexity of biomembranes, as well as by the limited tools for membrane lipidome manipulation in vivo. My PhD project was designed to bridge this gap through exploring how lipidome size and complexity define cell membrane organization and bioactivity. To achieve the overarching goal, we have created a minimal membrane model system using one of the simplest self-replicating organisms – the mammalian pathogen Mycoplasma. By taking advantage of Mycoplasma’s limited biosynthetic machinery we have established a method of controlling membrane lipid composition in M. mycoides and its synthetic counterpart JCVI Syn3, and have showed that we can finely tune Mycoplasma membrane lipidome through lipid diet. By utilizing our membrane model system in our manuscript, we showed that Mycoplasma membranes are forced to undergo larger magnitudes of lipidomic remodeling in response to temperature when cultivated on lipid diets with a smaller variety of lipid species. We have also demonstrated that larger lipidomes support homeoviscous adaptation in Mycoplasma. Overall, our study showed that varying lipidome size changes the efficiency of membrane lipidome remodeling in response to environmental changes and affects cellular ability to adapt its biophysical properties. At the later stage of my PhD, we have analyzed Mycoplasma transcriptomes to explore protein components responsible for membrane assembly and regulation as a function of membrane lipid composition. The preliminary results identified a non-ch

Osteochondral defects comprise cartilage and bone tissue in the joint region and create challenges for orthopedic surgery, also because intrinsic regeneration capacities of the articular cartilage are limited. Furthermore, tissue layer-specific characteristics regarding cell types, mechanical properties and biochemical composition need to be considered. Research questions: In this work, concepts were developed which allow mimicking of osteochondral interfacial layers in a patient-individual and zonally specified manner by 3D extrusion (bio)printing. This feature of patient specificity was proven on different levels within this project: Besides the option for application of patient-own, expanded stem cells or chondrocytes within a scaffold to support regeneration and neo-tissue formation, a workflow was implemented which enables the consideration of magnetic resonance imaging (MRI) data and zonal geometry of the defect. With the materials suitable to achieve this design and a bioprinting-compatible process, the impact of such a system on embedded cells was investigated. A zonally structured, partly mineralized construct was evaluated regarding its capability to allow or support chondrogenesis of primary human chondrocytes (hChon). Furthermore, a strategy based on core-shell bioprinting technology was developed which allows simultaneous embedding of different cell types in a zonally defined distribution with a targeted effect by incorporated growth factors while reducing the off-target effects that would be expected when applied homogeneously via the surrounding medium. In addition, hybrid multi-material scaffolds were developed to adjust the stiffness of these systems. Materials and methods: To define design and patient-specific requirements for an osteochondral implant, an anonymized MRI dataset of a patient with osteochondritis dissecans (OCD) was used. The main constituent of the developed fabrication system was a bioink based on 3% alginate and 9% methylcellulose, Osteochondrale Defekte umfassen Knochen- und Knorpelgewebe innerhalb des betroffenen Gelenks und stellen die klinische Orthopädie vor Herausforderungen dar, auch da die intrinsische Regenerationsfähigkeit des Gelenkknorpels stark limitiert ist. Zudem sind in den zu unterscheidenden Gewebeschichten spezifische Charakteristika wie unterschiedliche Zelltypen, mechanische Eigenschaften und die biochemische Zusammensetzung zu berücksichtigen. Fragestellungen: In der vorliegenden Arbeit wurden Konzepte entwickelt, mit dem sich per 3D-Extrusions(bio)druck Gewebeschichten dieser osteochondralen Grenzschicht zonenspezifisch und patientenindividuell nachbilden lassen. Diese patientenindividuellen Merkmale wurden innerhalb des Projektes auf mehreren Ebenen nachgewiesen: Zum einen können patienteneigene Stammzellen oder Chondrozyten nach Vermehrung im Labor innerhalb einer Gerüststruktur (“Scaffold”) zur Unterstützung der Regeneration und Gewebeneubildung angewandt werden. Zum anderen wurde ein Workflow vorgestellt, der die Berücksichtigung einer individuellen, per Magnetresonanztomographie (MRT) detektierten, schichtweisen Geometrie einer Läsion erlaubt. Mit Hilfe von Materialien, die diese Formgebung ermöglichen, wurde in einem Biodruck-kompatiblen Prozess der Einfluss eines solchen Systems auf eingebettete Zellen untersucht: Ein zonal aufgebautes, teilweise mineralisiertes Konstrukt wurde hinsichtlich dessen Eignung, Chondrogenese humaner Knorpelzellen (hChon) zu ermöglichen oder zu unterstützen, evaluiert. Zudem wurde eine auf der Kern-Mantel-Biodrucktechnologie basierende Strategie entwickelt, die das Einbetten unterschiedlicher Zelltypen mit zonal definierter Verteilung kombiniert mit einem gezielten Effekt durch inkorporierte Wachstumsfaktoren. Hierbei sollten unerwünschte Nebeneffekte der im Kern dargebrachten Faktoren auf die jeweils andere Zellsorte, die man bei homogener Faktorengabe über das umgebende Medium erwarten würde, reduziert werden. Weiterhin sollte mittels

Purpose – This study aims to identify the counterproductive knowledge behavior (CKB) of volunteers in nonprofit organizations and its influencing factors, based on the theories of planned behavior andwell-being. Design/methodology/approach – An online survey was used to collect 496 valid responses. A structural equation model was constructed, and the relationships among the constructs were estimated via the maximum likelihood method. To analyze the direct and indirect effects, 2,000 bootstrapping runs were conducted. A Kruskal-Wallis testwas also conducted to analyze the relationship between the variables. Findings – A combination of organizational factors and individual attitudes and perceptions can be used to explain CKB. Insecurity about knowledge sharing had the greatest impact on CKB. A competitive organizational norm induced CKB while a knowledge-sharing organizational norm did not have a significant impact. Further, the more self-determined the volunteer activity was, the more the CKB was suppressed. However, well-being did not have a significant direct effect. Volunteers with high levels of well-being and self-determination had significantly lower levels of insecurity about knowledge sharing compared to those who did not. Practical implications – Well-being arising fromvolunteering did not directly suppress CKB. To improve organizational efficiency by reducing CKB, nonprofit organization managers should provide intrinsically motivating tasks and interact with the volunteers. Originality/value – There is a lack of empirical research on CKB in volunteer organizations; therefore, the authors propose a new approach to knowledge management in volunteer activities., This work was supported by JSPS KAKENHI Grant Number 20H01529 and JAIST Open Access Paper Support Project.

In this article, we introduce the solution we used in the VSPW 2021 Challenge. Our experiments are based on two baseline models, Swin Transformer and MaskFormer. To further boost performance, we adopt stochastic weight averaging technique and design hierarchical ensemble strategy. Without using any external semantic segmentation dataset, our solution ranked the 5th place in the private leaderboard. Besides, we have some interesting attempts to tackle long-tail recognition and overfitting issues, which achieves improvement on val subset. Maybe due to distribution difference, these attempts don't work on test subset. We will also introduce these attempts and hope to inspire other researchers., Comment: Presented in ICCV'21 Workshop

Contrastive learning can largely enhance the feature discriminability in a self-supervised manner and has achieved remarkable success for various visual tasks. However, it is undesirably observed that the standard contrastive paradigm (features+$\ell_{2}$ normalization) only brings little help for domain adaptation. In this work, we delve into this phenomenon and find that the main reason is due to the class weights (weights of the final fully connected layer) which are vital for the recognition yet ignored in the optimization. To tackle this issue, we propose a simple yet powerful Probabilistic Contrastive Learning (PCL), which does not only assist in extracting discriminative features but also enforces them to be clustered around the class weights. Specifically, we break the standard contrastive paradigm by removing $\ell_{2}$ normalization and replacing the features with probabilities. In this way, PCL can enforce the probability to approximate the one-hot form, thereby reducing the deviation between the features and class weights. Benefiting from the conciseness, PCL can be well generalized to different settings. In this work, we conduct extensive experiments on five tasks and observe consistent performance gains, i.e., Unsupervised Domain Adaptation (UDA), Semi-Supervised Domain Adaptation (SSDA), Semi-Supervised Learning (SSL), UDA Detection, and UDA Semantic Segmentation. Notably, for UDA Semantic Segmentation on SYNTHIA, PCL surpasses the sophisticated CPSL-D by $>\!2\%$ in terms of mean IoU with a much smaller training cost (PCL: 1*3090, 5 days v.s. CPSL-D: 4*V100, 11 days). Code is available at https://github.com/ljjcoder/Probabilistic-Contrastive-Learning.

Performing autonomous exploration is essential for unmanned aerial vehicles (UAVs) operating in unknown environments. Often, these missions start with building a map for the environment via pure exploration and subsequently using (i.e. exploiting) the generated map for downstream navigation tasks. Accomplishing these navigation tasks in two separate steps is not always possible or even disadvantageous for UAVs deployed in outdoor and dynamically changing environments. Current exploration approaches either use a priori human-generated maps or use heuristics such as frontier-based exploration. Other approaches use learning but focus only on learning policies for specific tasks by either using sample inefficient random exploration or by making impractical assumptions about full map availability. In this paper, we develop an adaptive exploration approach to trade off between exploration and exploitation in one single step for UAVs searching for areas of interest (AoIs) in unknown environments using Deep Reinforcement Learning (DRL). The proposed approach uses a map segmentation technique to decompose the environment map into smaller, tractable maps. Then, a simple information gain function is repeatedly computed to determine the best target region to search during each iteration of the process. DDQN and A2C algorithms are extended with a stack of LSTM layers and trained to generate optimal policies for the exploration and exploitation, respectively. We tested our approach in 3 different tasks against 4 baselines. The results demonstrate that our proposed approach is capable of navigating through randomly generated environments and covering more AoI in less time steps compared to the baselines.

Nonviral gene delivery, though limited by inefficiency, has extensive utility in cell therapy, tissue engineering, and diagnostics. Substrate-mediated gene delivery (SMD) increases efficiency and allows transfection at a cell-biomaterial interface, by immobilizing and concentrating nucleic acid complexes on a surface. Efficient SMD generally requires substrates to be coated with serum or other protein coatings to mediate nucleic acid complex immobilization, as well as cell adhesion and growth; however, this strategy limits reproducibility and may be difficult to translate for clinical applications. As an alternative, we screened a chemically defined combinatorial library of 20 different extracellular matrix mimetic substrates containing combinations of (1) different sulfated polysaccharides that are essential extracellular matrix glycosaminoglycans (GAGs), with (2) mimetic peptides derived from adhesion proteins, growth factors, and cell-penetrating domains, for use as SMD coatings. We identified optimal substrates for DNA lipoplex and polyplex SMD transfection of fibroblasts and human mesenchymal stem cells. Optimal extracellular matrix mimetic substrates varied between cell type, donor source, and transfection reagent, but typically contained Heparin GAG and an adhesion peptide. Multiple substrates significantly increased transgene expression (i.e. 2- to 20-fold) over standard protein coatings. Considering previous research of similar ligands, we hypothesize extracellular matrix mimetic substrates modulate cell adhesion, proliferation, and survival, as well as plasmid internalization and trafficking. Our results demonstrate the utility of screening combinatorial extracellular matrix mimetic substrates for optimal SMD transfection towards application- and patient-specific technologies.

Xiaoyi Zhang, 1, 2 Yixin Zhang, 1, 2 Yaonan Wang, 1, 2 Jianhui Wu, 1, 2 Haiyan Chen, 1, 2 Ming Zhao, 1–3 Shiqi Peng 1, 2 1Beijing Area Major Laboratory of Peptide and Small Molecular Drugs, Department of Medicinal Chemistry, School of Pharmaceutical Sciences, Capital Medical University, Beijing 100069, People’s Republic of China; 2Engineering Research Center of Endogenous Prophylactic of Ministry of Education of China, Department of Medicinal Chemistry, School of Pharmaceutical Sciences, Capital Medical University, Beijing 100069, People’s Republic of China; 3Beijing Laboratory of Biomedical Materials and Key Laboratory of Biomedical Materials of Natural Macromolecules, Department of Biomaterials, College of Materials Science and Engineering, Beijing University of Chemical Technology, Beijing 100026, People’s Republic of ChinaCorrespondence: Shiqi Peng; Ming ZhaoDepartment of Medicinal Chemistry, School of Pharmaceutical Sciences, Capital Medical University, No. 10, Youanmenwaixitoutiao, Fengtai District, Beijing 100069, People’s Republic of ChinaTel +86 10 8391 1528; +86 10 8391 1535Fax +86 10 8391 1528; +86 10 8391 1533Email sqpeng@bjmu.edu.cn; maozhao@126.comBackground: 1-(4-isopropylphenyl)-β-carboline-3-carboxylic acid (ICCA) was modified by Trp-Phe-Phe to form 1-(4-isopropylphenyl)-β-carboline-3-carbonyl-Trp-Phe-Phe (ICCA-WFF).Purpose: The object of preparing ICCA-WFF was to enhance the in vivo efficacy of ICCA, to explore the possible targeting action, and to visualize the nano-feature.Methods: The advantages of ICCA-WFF over ICCA were demonstrated by a series of in vivo assays, such as anti-tumor assay, anti-arterial thrombosis assay, anti-venous thrombosis assay, P-selectin expression assay, and GPIIb/IIIa expression assay. The nano-features of ICCA-WFF were visualized by TEM, SEM and AFM images. The thrombus targeting and tumor-targeting actions were evidenced by FT-MS spectrum analysis.Re

We propose and experimentally demonstrate a technique using a composite-double-probe-pulse (CDPP) to eliminate the effect of polarization fading for phase-sensitive optical time-domain reflectometry (Φ-OTDR) based on ultra-weak FBG (UWFBG) array. The CDPP is composed of two optical pulses whose spatial interval is equal to twice the spatial interval of adjacent UWFBGs in the UWFBG array. One optical pulse is a long optical pulse, and the other optical pulse is composed of two continuous short optical pulses, whose polarization states are orthogonal to each other. The width of the short pulse is equal to half of the width of the normal pulse and their frequencies are different from the long pulse. By using such a method to perform the sensing for the UWFBG array, distributed quantitative measurement can be realized with only direct detection scheme and the influence of polarization fading in the demodulation of signal is thoroughly eliminated.

River ecosystems receive and process vast quantities of terrestrial organic carbon, the fate of which depends strongly on microbial activity. Variation in and controls of processing rates, however, are poorly characterized at the global scale. In response, we used a peer-sourced research network and a highly standardized carbon processing assay to conduct a global-scale field experiment in greater than 1000 river and riparian sites. We found that Earth's biomes have distinct carbon processing signatures. Slow processing is evident across latitudes, whereas rapid rates are restricted to lower latitudes. Both the mean rate and variability decline with latitude, suggesting temperature constraints toward the poles and greater roles for other environmental drivers (e.g., nutrient loading) toward the equator. These results and data set the stage for unprecedented "next-generation biomonitoring" by establishing baselines to help quantify environmental impacts to the functioning of ecosystems at a global scale.

River ecosystems receive and process vast quantities of terrestrial organic carbon, the fate of which depends strongly on microbial activity. Variation in and controls of processing rates, however, are poorly characterized at the global scale. In response, we used a peer-sourced research network and a highly standardized carbon processing assay to conduct a global-scale field experiment in greater than 1000 river and riparian sites. We found that Earth’s biomes have distinct carbon processing signatures. Slow processing is evident across latitudes, whereas rapid rates are restricted to lower latitudes. Both the mean rate and variability decline with latitude, suggesting temperature constraints toward the poles and greater roles for other environmental drivers (e.g., nutrient loading) toward the equator. These results and data set the stage for unprecedented “next-generation biomonitoring” by establishing baselines to help quantify environmental impacts to the functioning of ecosystems at a global scale.