RESUMEN
Graph-based semi-supervised learning methods have been used in a wide range of real-world applications, e.g., from social relationship mining to multimedia classification and retrieval. However, existing methods are limited along with high computational complexity or not facilitating incremental learning, which may not be powerful to deal with large-scale data, whose scale may continuously increase, in real world. This paper proposes a new method called Data Distribution Based Graph Learning (DDGL) for semi-supervised learning on large-scale data. This method can achieve a fast and effective label propagation and supports incremental learning. The key motivation is to propagate the labels along smaller-scale data distribution model parameters, rather than directly dealing with the raw data as previous methods, which accelerate the data propagation significantly. It also improves the prediction accuracy since the loss of structure information can be alleviated in this way. To enable incremental learning, we propose an adaptive graph updating strategy which can update the model when there is distribution bias between new data and the already seen data. We have conducted comprehensive experiments on multiple datasets with sample sizes increasing from seven thousand to five million. Experimental results on the classification task on large-scale data demonstrate that our proposed DDGL method improves the classification accuracy by a large margin while consuming much less time compared to state-of-the-art methods.
RESUMEN
Patient survival prediction based on gigapixel whole-slide histopathological images (WSIs) has become increasingly prevalent in recent years. A key challenge of this task is achieving an informative survival-specific global representation from those WSIs with highly complicated data correlation. This article proposes a multi-hypergraph based learning framework, called "HGSurvNet," to tackle this challenge. HGSurvNet achieves an effective high-order global representation of WSIs via multilateral correlation modeling in multiple spaces and a general hypergraph convolution network. It has the ability to alleviate over-fitting issues caused by the lack of training data by using a new convolution structure called hypergraph max-mask convolution. Extensive validation experiments were conducted on three widely-used carcinoma datasets: Lung Squamous Cell Carcinoma (LUSC), Glioblastoma Multiforme (GBM), and National Lung Screening Trial (NLST). Quantitative analysis demonstrated that the proposed method consistently outperforms state-of-the-art methods, coupled with the Bayesian Concordance Readjust loss. We also demonstrate the individual effectiveness of each module of the proposed framework and its application potential for pathology diagnosis and reporting empowered by its interpretability potential.
Asunto(s)
Algoritmos , Aprendizaje , Humanos , Teorema de BayesRESUMEN
Sketch-based image retrieval (SBIR) is a long-standing research topic in computer vision. Existing methods mainly focus on category-level or instance-level image retrieval. This paper investigates the fine-grained scene-level SBIR problem where a free-hand sketch depicting a scene is used to retrieve desired images. This problem is useful yet challenging mainly because of two entangled facts: 1) achieving an effective representation of the input query data and scene-level images is difficult as it requires to model the information across multiple modalities such as object layout, relative size and visual appearances, and 2) there is a great domain gap between the query sketch input and target images. We present SceneSketcher-v2, a Graph Convolutional Network (GCN) based architecture to address these challenges. SceneSketcher-v2 employs a carefully designed graph convolution network to fuse the multi-modality information in the query sketch and target images and uses a triplet training process and end-to-end training manner to alleviate the domain gap. Extensive experiments demonstrate SceneSketcher-v2 outperforms state-of-the-art scene-level SBIR models with a significant margin.
RESUMEN
Background: Coronavirus Disease-19 (COVID-19), a rising global pandemic, has triggered psychological crises among the public. Panic, a severe symptom of mental disorders, is increasing in the public in China and it is urgent to provide research for intervention development. Objectives: This study aimed to assess the prevalence of public panic in China during the earliest stage of the COVID-19 pandemic and to explore the associated psychological behavioral responses and public's risk perception of the pandemic. Methods: A cross-sectional study using a web-based survey with convenience sampling was conducted with 2,484 participants nationally from February 11 to February 24, 2020 in China. A self-developed questionnaire was applied to assess the prevalence of public panic and its associated factors. Multivariable logistic regression analysis was applied to assess the risk and protective factors of public panic. Results: There were 23.39% (581/2,484) of the participants who reported experiencing panic during the earliest stage of the COVID-19 pandemic. Taking temperature repeatedly, being nervous in a crowd, being suspicious of infection in the family, being worried about the future, and worries about high infectivity of the COVID-19, lack of effective therapies, and wide impact of the COVID-19 pandemic increased the odds of public panic. Whereas, avoiding gatherings during holidays was negatively associated with the odds of public panic. Conclusions: Psycho-behavioral responses were closely associated with public panic during the earliest stage of the COVID-19 pandemic in China. Defusing excessive health-related worries, the guidance of appropriate self-protective behaviors, strengthening of health education in communities, and available treatment for mental disorders should be adopted to monitor the psychological responses and to guide the behaviors of the public.
RESUMEN
Hypergraph learning is a technique for conducting learning on a hypergraph structure. In recent years, hypergraph learning has attracted increasing attention due to its flexibility and capability in modeling complex data correlation. In this paper, we first systematically review existing literature regarding hypergraph generation, including distance-based, representation-based, attribute-based, and network-based approaches. Then, we introduce the existing learning methods on a hypergraph, including transductive hypergraph learning, inductive hypergraph learning, hypergraph structure updating, and multi-modal hypergraph learning. After that, we present a tensor-based dynamic hypergraph representation and learning framework that can effectively describe high-order correlation in a hypergraph. To study the effectiveness and efficiency of hypergraph generation and learning methods, we conduct comprehensive evaluations on several typical applications, including object and action recognition, Microblog sentiment prediction, and clustering. In addition, we contribute a hypergraph learning development toolkit called THU-HyperG.
RESUMEN
Link prediction aims at inferring missing links or predicting future ones based on the currently observed network. This topic is important for many applications such as social media, bioinformatics and recommendation systems. Most existing methods focus on homogeneous settings and consider only low-order pairwise relations while ignoring either the heterogeneity or high-order complex relations among different types of nodes, which tends to lead to a sub-optimal embedding result. This paper presents a method named Heterogeneous Hypergraph Variational Autoencoder (HeteHG-VAE) for link prediction in heterogeneous information networks (HINs). It first maps a conventional HIN to a heterogeneous hypergraph with a certain kind of semantics to capture both the high-order semantics and complex relations among nodes, while preserving the low-order pairwise topology information of the original HIN. Then, deep latent representations of nodes and hyperedges are learned by a Bayesian deep generative framework from the heterogeneous hypergraph in an unsupervised manner. Moreover, a hyperedge attention module is designed to learn the importance of different types of nodes in each hyperedge. The major merit of HeteHG-VAE lies in its ability of modeling multi-level relations in heterogeneous settings. Extensive experiments on real-world datasets demonstrate the effectiveness and efficiency of the proposed method.
RESUMEN
Reconstructing a 3D shape from a single-view image using deep learning has become increasingly popular recently. Most existing methods only focus on reconstructing the 3D shape geometry based on image constraints. The lack of explicit modeling of structure relations among shape parts yields low-quality reconstruction results for structure-rich man-made shapes. In addition, conventional 2D-3D joint embedding architecture for image-based 3D shape reconstruction often omits the specific view information from the given image, which may lead to degraded geometry and structure reconstruction. We address these problems by introducing VGSNet, an encoder-decoder architecture for view-aware joint geometry and structure learning. The key idea is to jointly learn a multimodal feature representation of 2D image, 3D shape geometry and structure so that both geometry and structure details can be reconstructed from a single-view image. To this end, we explicitly represent 3D shape structures as part relations and employ image supervision to guide the geometry and structure reconstruction. Trained with pairs of view-aligned images and 3D shapes, the VGSNet implicitly encodes the view-aware shape information in the latent feature space. Qualitative and quantitative comparisons with the state-of-the-art baseline methods as well as ablation studies demonstrate the effectiveness of the VGSNet for structure-aware single-view 3D shape reconstruction.
RESUMEN
PURPOSE: Caregivers of stroke inpatients are at high risk of sleep disorder, which may lead to depressive symptoms. Self-efficacy has always been regarded as a protective factor against psychological disorders such as depressive symptoms. This study aims to investigate the sleep disorder and depressive symptoms of caregivers of stroke inpatients in China and explore the mediating effect of self-efficacy between sleep disorder and depressive symptoms among Chinese caregivers of stroke inpatients. PATIENTS AND METHODS: In this cross-sectional study, a total of 305 caregivers who were hospitalized with stroke patients completed the PROMIS Sleep Disorder Short Form Scale, General Self-Efficacy Scale and Patient Health Questionnaire-9 in two general public hospitals in northeast and southeast China. A structural equation model with bootstrap method was performed to determine the mediation of self-efficacy between sleep disorder and depressive symptoms. RESULTS: Among the participants, 55.4% of caregivers reported depressive symptoms. Sleep disorder and self-efficacy were significant predictors of depressive symptoms. The direct impact of sleep disorder on depressive symptoms was positive, and the path coefficient of sleep disorder with depressive symptoms was decreased from 0.45 to 0.38 (P < 0.01) after addition of self-efficacy in the model. This indicated that self-efficacy played as mediator. CONCLUSION: The caregivers of stroke inpatients were in poor physical and psychological health, and more than half of the caregivers (55.4%) suffered from depressive symptoms. Our research revealed the mediation of self-efficacy between sleep disorder and depressive symptoms, and emphasized the importance of enhancing self-efficacy to reduce depressive symptoms among caregivers of stroke inpatients. These results demonstrate that focusing on self-efficacy interventions can enhance mental health and reduce depressive symptoms effectively.
RESUMEN
The continuous emergence of drug-target interaction data provides an opportunity to construct a biological network for systematically discovering unknown interactions. However, this is challenging due to complex and heterogeneous correlations between drug and target. Here, we describe a heterogeneous hypergraph-based framework for drug-target interaction (HHDTI) predictions by modeling biological networks through a hypergraph, where each vertex represents a drug or a target and a hyperedge indicates existing similar interactions or associations between the connected vertices. The hypergraph is then trained to generate suitably structured embeddings for discovering unknown interactions. Comprehensive experiments performed on four public datasets demonstrate that HHDTI achieves significant and consistently improved predictions compared with state-of-the-art methods. Our analysis indicates that this superior performance is due to the ability to integrate heterogeneous high-order information from the hypergraph learning. These results suggest that HHDTI is a scalable and practical tool for uncovering novel drug-target interactions.
RESUMEN
Three-dimensional face dense alignment and reconstruction in the wild is a challenging problem as partial facial information is commonly missing in occluded and large pose face images. Large head pose variations also increase the solution space and make the modeling more difficult. Our key idea is to model occlusion and pose to decompose this challenging task into several relatively more manageable subtasks. To this end, we propose an end-to-end framework, termed as Self-aligned Dual face Regression Network (SADRNet), which predicts a pose-dependent face, a pose-independent face. They are combined by an occlusion-aware self-alignment to generate the final 3D face. Extensive experiments on two popular benchmarks, AFLW2000-3D and Florence, demonstrate that the proposed method achieves significant superior performance over existing state-of-the-art methods.
RESUMEN
BACKGROUND: Prenatal anxiety is one of the most prevalent mental disorders during pregnancy. This study assessed the prevalence of prenatal anxiety and examined whether resilience could play the mediating role in the association between self-efficacy and symptoms of prenatal anxiety among pregnant women in China. METHODS: A nationwide smartphone cross-sectional study was carried out in three cities (Shenyang of Liaoning Province, Zhengzhou of Henan Province and Chongqing Municipality) in China from July 2018 to July 2019. The questionnaire consisted of questions on demographic characteristics, the Generalized Anxiety Disorder Scale (GAD-7), the Chinese version of General Self-efficacy Scale (GSES), and the 14-item Wagnild and Young Resilience Scale (RS-14). A total of 665 pregnant women were recruited in this study. A hierarchical multiple regression model was employed to explore the associate factors and mediators of symptoms of prenatal anxiety. A structural equation model was employed to test the hypothesis that resilience mediates the association between self-efficacy and symptoms of prenatal anxiety. RESULTS: The prevalence of symptoms of prenatal anxiety was 36.4% in this study. Self-efficacy was negatively correlated with symptoms of prenatal anxiety (r = -0.366, P < 0.01). Resilience had a significant positive correlation with self-efficacy (r = 0.612, P < 0.01) and had a negative correlation with symptoms of prenatal anxiety (r = -0.427, P < 0.01). The hierarchical multiple regression model indicated that self-efficacy and resilience were the main factors associated with symptoms of prenatal anxiety and contributed to 11.9% and 6.3% to the variance of symptoms of prenatal anxiety, respectively. Resilience served as a mediator between self-efficacy and symptoms of prenatal anxiety (a*b = -0.198, Bias-corrected and accelerated bootstrap 95% Confidence interval: -0.270, -0.126). CONCLUSIONS: Self-efficacy was a negative predictor of symptoms of prenatal anxiety among pregnant women. Moreover, resilience mediated the relation between self-efficacy and symptoms of prenatal anxiety among pregnant women in China. It was observed in this study that psychological interventions might be beneficial for pregnant women to relieve symptoms of prenatal anxiety through improved self-efficacy and resilience.
Asunto(s)
Ansiedad/psicología , Mujeres Embarazadas/psicología , Resiliencia Psicológica , Autoeficacia , Adulto , China/epidemiología , Estudios Transversales , Femenino , Humanos , Embarazo , Prevalencia , Teléfono Inteligente , Encuestas y CuestionariosRESUMEN
Background: Stroke patients may suffer from a variety of symptoms which can result in sleep disturbance and post-stroke depression (PSD). Whereas, resilience can alleviate sleep disturbance and help maintain well-being after stroke. Objective: The aim of this study is to explore whether resilience plays a mediating role in the relationship between sleep disturbance and PSD of stroke patients in China. Methods: A cross-sectional study with a multi-stage sampling was carried out in Liaoning Rehabilitation Center and the Third People's Hospital of Chongqing in China from May to September 2019. A total of 353 stroke patients were enrolled in this study. Structural equation model (SEM) was used to test the mediating effect of resilience on the relationship between sleep disturbance and PSD. Results: The prevalence of PSD of stroke patients was 34.56%. Sleep disturbance contributed most to the variance of PSD and had a significantly positive association with PSD among stroke patients (P < 0.01). Resilience was negatively associated with PSD, and acted as a mediator between sleep disturbance and PSD (a * b = 0.201, BCa 95% CI: 0.156~0.254). Conclusions: The prevalence of PSD was high among the Chinese stroke patients. Sleep disturbance was highly associated with PSD, resulting in the increased risk of PSD. Furthermore, resilience has a mediating effect on the relationship between sleep disturbance and PSD, and could reduce the negative effect of sleep disturbance on the development of PSD.
RESUMEN
BACKGROUND: The rapid outbreak of COVID-19 around the world has adversely affected the mental health of the public. The prevalence of anxiety among the public has increased dramatically during the COVID-19 pandemic. However, there are few studies evaluating the effects of positive psychological responses and information-seeking behaviors on anxiety experienced among social media users during the COVID-19 pandemic. OBJECTIVE: This study evaluated the prevalence of anxiety and its associated factors among WeChat users in mainland China during the early stages of the COVID-19 pandemic. METHODS: From February 10 to February 24, 2020, a nationwide, web-based cross-sectional survey study was carried out using convenience sampling. Participants' levels of anxiety, positive psychological responses, and information-seeking behaviors were assessed. The survey was distributed among WeChat users via the WeChat smartphone platform. Chi-square tests and multivariable logistic regression analyses were performed to examine the factors associated with anxiety. RESULTS: This study found that the prevalence of anxiety (Generalized Anxiety Disorder 7-item [GAD-7] scale score ≥7) among WeChat users in China was 17.96% (446/2483) during the early stages of the COVID-19 pandemic. Results of multivariable logistic regression analysis showed that information-seeking behaviors such as cannot stop searching for information on COVID-19, being concerned about the COVID-19 pandemic, and spending more than 1 hour per day consuming information about the pandemic were found to be associated with increased levels of anxiety. Additionally, participants who chose social media and commercial media as the primary sources to obtain information about the COVID-19 pandemic were found more likely to report anxiety. Conversely, participants who were confident or rational about the COVID-19 pandemic were less likely to report anxiety. CONCLUSIONS: This study found that positive psychological responses and information-seeking behaviors were closely associated with anxiety among WeChat users during the COVID-19 pandemic in China. It might be paramount to enhance mental well-being by helping people respond to the COVID-19 pandemic more rationally and positively in order to decrease symptoms of anxiety.
Asunto(s)
Ansiedad/epidemiología , COVID-19/epidemiología , COVID-19/psicología , Medios de Comunicación Sociales/estadística & datos numéricos , Adulto , Ansiedad/etiología , Ansiedad/psicología , China/epidemiología , Estudios Transversales , Brotes de Enfermedades , Femenino , Humanos , Masculino , Pandemias , SARS-CoV-2/aislamiento & purificación , Encuestas y CuestionariosRESUMEN
Sketches in existing large-scale datasets like the recent QuickDraw collection are often stored in a vector format, with strokes consisting of sequentially sampled points. However, most existing sketch recognition methods rasterize vector sketches as binary images and then adopt image classification techniques. In this article, we propose a novel end-to-end single-branch network architecture RNN-Rasterization-CNN (Sketch-R2CNN for short) to fully leverage the vector format of sketches for recognition. Sketch-R2CNN takes a vector sketch as input and uses an RNN for extracting per-point features in the vector space. We then develop a neural line rasterization module to convert the vector sketch and the per-point features to multi-channel point feature maps, which are subsequently fed to a CNN for extracting convolutional features in the pixel space. Our neural line rasterization module is designed in a differentiable way for end-to-end learning. We perform experiments on existing large-scale sketch recognition datasets and show that the RNN-Rasterization design brings consistent improvement over CNN baselines and that Sketch-R2CNN substantially outperforms the state-of-the-art methods.
RESUMEN
Three-dimensional multi-modal data are used to represent 3D objects in the real world in different ways. Features separately extracted from multimodality data are often poorly correlated. Recent solutions leveraging the attention mechanism to learn a joint-network for the fusion of multimodality features have weak generalization capability. In this paper, we propose a hamming embedding sensitivity network to address the problem of effectively fusing multimodality features. The proposed network called HamNet is the first end-to-end framework with the capacity to theoretically integrate data from all modalities with a unified architecture for 3D shape representation, which can be used for 3D shape retrieval and recognition. HamNet uses the feature concealment module to achieve effective deep feature fusion. The basic idea of the concealment module is to re-weight the features from each modality at an early stage with the hamming embedding of these modalities. The hamming embedding also provides an effective solution for fast retrieval tasks on a large scale dataset. We have evaluated the proposed method on the large-scale ModelNet40 dataset for the tasks of 3D shape classification, single modality and cross-modality retrieval. Comprehensive experiments and comparisons with state-of-the-art methods demonstrate that the proposed approach can achieve superior performance.