299645 research outputs found
Sort by
WRF-HEATS coupling: Incorporating human behaviors and city topography into urban heat stress evaluation
No full textUrban human thermal stress can be inaccurately estimated along with less-understood heat heterogeneity due to the absence of high-resolution meteorological information and realistic human behavior representation. To this end, we coupled a regional climate model (weather research and forecasting model, WRF) and a human energy balance model (human-environment adaptive thermal stress model, HEATS) to predict pedestrian's dynamic thermal stress at a neighborhood scale. The WRF-human coupling system resolves human-environment heat exchanges based on meteorological and topographical information with the consideration of dynamic human activities. The coupling system has been tested and utilized to study dynamic heat stress in a typical hot, humid, and mountainous city, Hong Kong. Our results revealed widespread heat heterogeneity with up to 7 °C difference in Physiological Subjective Temperature (PST) in the core urban area, and extreme heat exposure (up to 45 °C PST) in calm-wind zones at noon. Heat stress can be further aggregated considering realistic human behaviors such as extra clothing (e.g., protective facemask during pandemics) and physical exercise (e.g., walking along inclined terrain). Optimal-thermal-comfort routes have been designed and suggested based on the simulated neighborhood-scale heat stress map. </p
Treatment futility: Continuation or withdrawal of life-sustaining treatment in intensive care units
No full textGenerative Adversarial Networks for Imputing Sparse Learning Performance
No full textLearning performance data, such as correct or incorrect responses to questions in Intelligent Tutoring Systems (ITSs) is crucial for tracking and assessing the learners’ progress and mastery of knowledge. However, the issue of data sparsity, characterized by unexplored questions and missing attempts, hampers accurate assessment and the provision of tailored, personalized instruction within ITSs. This paper proposes using the Generative Adversarial Imputation Networks (GAIN) framework to impute sparse learning performance data, reconstructed into a three-dimensional (3D) tensor representation across the dimensions of learners, questions and attempts. Our customized GAIN-based method computational process imputes sparse data in a 3D tensor space, significantly enhanced by convolutional neural networks for its input and output layers. This adaptation also includes the use of a least squares loss function for optimization and aligns the shapes of the input and output with the dimensions of the questions-attempts matrices along the learners’ dimension. Through extensive experiments on six datasets from various ITSs, including AutoTutor, ASSISTments and MATHia, we demonstrate that the GAIN approach generally outperforms existing methods such as tensor factorization and other generative adversarial network (GAN) based approaches in terms of imputation accuracy. This finding enhances comprehensive learning data modeling and analytics in AI-based education.link_to_subscribed_fulltex
The development of community-based smoking cessation interventions in Hong Kong
No full textTobacco kills up to half of its users[1−3], yet one-fifth of the world's population is currently smoking[4]. The World Health Organization (WHO) has launched the Framework Convention on Tobacco Control (FCTC)[5] and the MPOWER policy package (Monitor, Protect, Offer, Warn, Enforce, Raise)[1] to guide country-level implementation of effective smoking cessation measures. Smoking cessation at any age has significant immediate and long-term health benefits[6], and 'Offer help to quit tobacco use' is one of the key components in MPOWER tobacco control strategy, highlighting the crucial role of primary health services in providing smoking cessation services[7,8].Since 1982, Hong Kong has progressively integrated the legislation and enforcement, taxation, publicity and education, and smoking cessation services as a comprehensive tobacco control strategy[9]. The multi-pronged approach has gradually reduced the smoking prevalence among the population aged 15 or above from 23.3% in 1982 to 10.1% in 2018, and 9.1% in 2023[10]. However, there is still an enormous smoking-related burden given the 637,900 daily cigarette smokers in Hong Kong and accounts for over 7,000 premature deaths per year. In 2018, the Hong Kong Department of Health launched an initiative to prevent and control non-communi-cable diseases 'Towards 2025' and targeted to cut the daily cigarette smoking prevalence from 11.1% in 2010 to 7.8% by 2025[11].Tobacco is highly addictive, and it is difficult for smokers with nicotine dependence to quit without assistance. Approximately two-thirds of adult smokers (68.0%) are willing to quit, but fewer than one-third of smokers who had previous quit attempts ever used evidence-based cessation treatments[12]. Further reduction in smoking prevalence is challenging as many remaining smokers are likely to be more 'hardcore' and thus find it more difficult to quit[13].Our participants of nine territory-wide smoking cessation randomized controlled trials from 2009 to 2018 (except 2011, N = 9,837) also found that cigarette smokers in Hong Kong were having less quit attempts, becoming less motivated to quit, and perceiving lower self-efficacy in quitting in the recent decade[14]. Smoking cessation treatments could substantially increase quit rate but were largely reactive[15]. As few smokers would actively seek help to quit, proactive recruitment is increasingly important to connect smokers to available effective smoking cessation services in the community[15,16]. Our and the research of others have achieved improvements in developing community-based smoking cessation interventions, but most of the current community tobacco control practices remain on preventing smoking initiation through educational programs. Through analyzing the development and effectiveness of community smoking cessation interventions, we aim to provide theoretical and practical references for the implementation of community smoking cessation intervention in the future.Community-based smoking cessation interventions mainly include pharmacological treatments to alleviate nicotine with-drawal symptoms and behavior interventions guided by social psychology models to enhance smoking cessation motivation and support smoking cessation attempts. Pharmacological treatmentinvolves the use of several approved drugs to assist in smoking cessation, mainly provided to smokers who are willing to make a quit attempt. Behavioural interventions vary widely in the content and delivery methods and can be passively (e.g., calling Quitline) or proactively (e.g., advising patients to quit during a visit) provided. They may be delivered to smokers who were unmotivated or not interested in quitting. Pharmacological treatments were usually conducted in conjunction with behavioural interventions (e.g.,counselling) to achieve improved cessation outcomes.</p
Association of blood cadmium and physical activity with mortality: A prospective cohort study
No full textPhysical activity (PA) may be considered an alternative method to ameliorate the elevated mortality risks associated with cadmium exposure. In this prospective cohort study, a total of 20,253 participants (weighted mean age, 47.79 years), including 10,247 men (weighted prevalence: 50.1 %), aged 18 years or older, were selected from the National Health and Nutrition Examination Survey from 2007 to 2018. Multivariable Cox proportional hazards regression models were utilized to evaluate the associations between blood cadmium levels, PA, and the risks of mortality. Restricted cubic spline analyses were employed to investigate the nonlinear relationships between blood cadmium and PA levels and mortality risks. During a median follow-up of 7.6 years, a total of 2002 (9.89 %) all-cause deaths occurred, of which 581 (2.87 %) participants were due to cardiovascular disease (CVD) and 498 (2.46 %) died of cancer. J-shaped associations were observed for blood cadmium with risks of mortality (all P overall < 0.001; all P nonlinearity < 0.001). Blood cadmium and PA had multiplicative interactions on mortality risk (all P interaction < 0.05). Compared with the subgroup with the lowest quartile of blood cadmium and recommended PA, the combination of the highest quartile of blood cadmium and without recommended PA was associated with the highest risks of all-cause and cancer mortality, followed by those meeting recommended PA but in the highest quartile of blood cadmium (hazard ratios, 2.43; 95 % confidence interval, 1.95–3.02). Achieving recommended PA significantly attenuated the detrimental effects of blood cadmium on all-cause, CVD, and cancer mortality risks
A Collaborative Model for Restorative Compensation in Public Interest Litigation Involving Aquatic Ecology in Guangdong Province, China
No full textThe Guangdong Province is rich in waterways, including those of the Pearl River. The entire watershed of the Pearl River system spans the territory of six provinces. Considering the overarching objective of building a ‘beautiful Bay Area’ under the guidance of Outline Development Plan for the Guangdong-Hong Kong-Macao Greater Bay Area as well as the ecological problems that span over river basins and regions in Guandong Province, public interest litigation is a useful tool in protecting the environment. Analyzing 95 first-instance (trial) judgements handed down in Guangdong Province between 2018 and 2021, we sought to evaluate public interest litigation as a means of safeguarding aquatic ecology in the Greater Bay Area (GBA), China. Cases were categorized for: firstly, their approach to determining the extent of ecological damage; secondly, the procedure used for receiving and auditing restorative compensation; thirdly, the collaboration between the court and government departments in the management and use of restorative compensation; and fourthly, the collaborative ‘public–private’ supervision utilized to monitor the implementation of restorative compensation and actual restoration. Our insights are intended to provide guidance for cooperative opportunities in the large transregional water systems and offshore areas of mainland China.</p
Advanced neural networks and their application for medical image segmentation
No full textConvolutional neural networks have shown significant promise in medical image segmentation, providing crucial insights for early diagnosis, biopsy planning, and clinical therapies. However, accurate segmentation is challenging due to inherent image characteristics (\textit{i.e.}, low contrast, speckle noise in ultrasound), variations in target shape and size, and domain shift across datasets. This thesis aims to develop advanced neural networks for medical image segmentation. Specifically, we shall consider breast tumor segmentation from ultrasound images, ultrasound thyroid nodule segmentation, and domain generalization in prostate MRI segmentation.
Firstly, we propose an innovative Multi-scale Dynamic Fusion Network (MDF-Net) to segment the ultrasound breast tumors. It is structured as an end-to-end two-stage architecture, comprising a trunk sub-network responsible for multi-scale feature selection and a refinement sub-network that is optimized to enhance feature exploration and fusion, thereby minimizing impairments. Building upon the UNet++, the trunk network features dense skip connections to facilitate connectivity between features across different scales. Additionally, we introduce multi-scale deep supervision to capture more discriminative features and attenuate inaccuracies stemming from speckle noise. The refinement sub-network leverages a structurally optimized MDF mechanism to enhance initial segmentation at coarser scales and delve into inter-subject variation insights at finer scales. Evaluation of two publicly available datasets demonstrates that our proposed MDF-Net outperforms state-of-the-art approaches in terms of Dice coefficient and other evaluation metrics.
Secondly, existing multi-task learning methods for thyroid nodule segmentation suffer from 1) the distribution gap between different datasets and 2) inconsistency in loss calculation for different tasks. We propose a novel STR-Net to address these issues. Specifically, we propose a Multi-mix Data Augmentation that randomly crops the foreground and background of gland and nodule images and mixes them to generate new samples. Furthermore, we propose a new Thyroid-Region Prior Guided Refinement Network by adding Multi-scale Deep Supervision and Nodule Refinement Structure. Moreover, a Teacher-Student Semi-supervised Framework is constructed with our proposed network to maintain consistency in the multi-task feature alignment. Finally, an Edge Distance Regularization method is proposed for post-processing to make nodule segmentation boundaries smoother and flatter. Extensive experiments on two datasets have demonstrated the effectiveness of our method.
Lastly, we propose a bidirectional Gated Recurrent Unit (GRU) based refinement network with simple and effective Patch Mixing and Risk Extrapolation (PMRE) schemes for multi-site prostate MRI segmentation. It employs a large convolution kernel-based multiscale feature encoder to extract multiscale features from consecutive 2D slices and a recurrent bidirectional ConvGRU-based contracting decoder to fuse the 3D segmentation features from a coarse-to-fine strategy. To enhance the generalization capability and robustness of the network across diverse target domains, a novel PMRE domain generalization approach is introduced by leveraging data manipulation, network design optimizations, and risk extrapolation. Specifically, an effective Domain Patch Mixing mechanism, which interpolates patches from different domains, is proposed for effective data augmentation. A simple and effective Segmentation Risk Extrapolation scheme is proposed to minimize the performance spread of the network over all the multisite samples. Experimental results from six commonly used source domains of prostate show that the proposed framework performs better than state-of-the-art algorithms.published_or_final_versionElectrical and Electronic EngineeringDoctoralDoctor of Philosoph
The role of IL-25 in the development of murine lupus
No full textSystemic lupus erythematosus (SLE) is an autoimmune disease characterized by multiple organ inflammatory damage and aberrant production of autoantibodies. Numerous studies have demonstrated significant roles of IL-17 and Th17 cells in the development of SLE. However, the roles of IL-25 (also named IL-17E), an important member of the IL-17 family cytokines, in the pathogenesis of SLE remain largely unclear. This study aims to investigate the role of IL-25 and elucidate the underlying effector mechanisms in the development of murine lupus.
In SLE patients, the concentrations of IL-25 in serum were markedly elevated compared with healthy donors and positively correlated with the SLE disease activity index (SLEDAI). Moreover, a murine lupus model was successfully established, which recapitulates the hallmark features of SLE patients, including renal damage, the presence of various autoantibodies, and excessive inflammatory responses. In chromatin-induced lupus mice, the levels of IL-25 in serum and kidney tissues were significantly increased compared with normal mice.
To investigate the role of IL-25 in lupus pathogenesis, wild-type (WT) and IL-25 knockout (KO) mice were immunized for lupus induction. Compared with WT counterparts, IL-25 KO mice exhibited significantly increased levels of autoantibodies and creatinine in serum as well as proteins in urine. Moreover, the histopathological analysis showed more severe renal tissue damage as evidenced by exacerbated kidney tissue inflammation and glomerular damage in IL-25 KO mice, suggesting that IL-25 protected lupus development in mice. Immune cell profiling revealed enhanced Th17 responses with increased numbers of GM-CSF+ pathogenic Th17 (pTh17) subsets in mice with IL-25 deficiency. In culture, IL-25 significantly suppressed differentiation of both mouse and human Th17 cells. Moreover, IL-25 decreased the expression of IL-23R and IL-1R, two key markers for pTh17 cells.
To elucidate the metabolic and molecular mechanism by which IL-25 regulates Th17 differentiation, cultured Th17 cells with and without IL-25 treatments were collected for a glycolysis stress test. IL-25 significantly decreased glycolysis levels in Th17 cells, together with reduced expression of glucose transporter 1 (GLUT1) and decreased glucose uptake. Gene expression analysis showed that IL-25 decreased the expression of many glycolysis-associated genes. In addition, IL-25 suppressed the expression of HIF-1α, a key glycolytic regulator. Mechanistically, IL-25 treatment inhibited phosphorylation of STAT3 and activation of AKT1-mTOR pathway, both of which play a critical role in modulating HIF-1α expression, glycolytic metabolism and Th17 responses.
Currently, effective therapies for SLE patients are still lacking. In this study, recombinant IL-25 was shown to effectively ameliorate lupus development in mice. The lupus mice exhibited reduced levels of proteinuria and serum creatinine upon IL-25 treatment. Histological assessments revealed that IL-25 administration markedly attenuated kidney damage. Moreover, IL-25 treatment significantly suppressed lupus development in a humanized lupus model established in NOD scid gamma (NSG) mice.
Taken together, these findings have revealed a regulatory role of IL-25 in lupus development by suppressing Th17 cell glycolysis and differentiation. Moreover, the preclinical study has identified IL-25 as a promising strategy for treating lupus in mice, which may contribute to the development of novel effective therapies for treating SLE patients.published_or_final_versionPathologyDoctoralDoctor of Philosoph
