Search

Background To investigate whether asymptomatic close-contact family members of patients diagnosed with coronavirus disease (COVID-19) should immediately undergo CT screening in addition to the viral nucleic acid test. Methods We retrospectively analyzed the data of a family cluster of 8 individuals, of whom 1 family member (Patient 3) had an epidemiologic history of having visited Guangzhou from Hubei Province on January 20, 2020. Her father (Patient 1) developed a fever and respiratory system symptoms and was confirmed COVID-19-positive on February 4-5, 2020 at Zengcheng People's Hospital, Guangzhou, China. Seven close-contact family members of the patients were then screened for COVID-19 on February 5-6 at the hospital. The CT imaging manifestation and laboratory tests of this family cluster were investigated and reported. Results Five (62.5%) of the 8 family members were confirmed COVID-19-positive. Except for Patient 1, who had fever, cough, fatigue, and dizziness, the remaining four (4/5, 80%) COVID-19-positive family members (Patients 2-5) had no clinical symptoms. Among the 5 patients, 2 had leukopenia (2/5, 40%), 1 had low absolute neutrophil counts (1/5, 20%), and 2 had increased high-sensitivity C-reactive protein (2/5, 40%). Ground-glass opacity (GGO) was found on chest CT imaging in all 5 patients (5/5, 100%), with interlobular septal thickening. Thickened blood vessel shadows were seen in 3 patients (3/5, 60%). The 3 COVID-19-negative family members (Family Members 1-3) did not have CT abnormalities, and they showed negative reverse transcription-polymerase chain reaction (RT-PCR) results twice. Conclusions CT screening is necessary in close-contact family members of a confirmed COVID-19 pneumonia case, regardless of the presence of clinical symptoms.

Recognizing human expression in videos is a challenging task due to dynamic changes in facial actions and diverse visual appearances. The key to design a reliable video-based expression recognition system is to extract robust spatial features and make full use of temporal modality characteristics. In this paper, we present a novel network architecture called Cascaded Attention Network (CAN) which is a cascaded spatiotemporal model incorporating with both spatial and temporal attention, tailored to video-level facial expression recognition. The cascaded fundamental model consists of a transfer convolutional network and Bidirectional Long Short-Term Memory (BiLSTM) network. Spatial attention is designed from the facial landmarks since facial expressions depend on the actions of key regions (eyebrows, eyes, nose, and mouth) on the face. Focusing on these key regions can help to decrease the effect of person-specific attributes. Meanwhile, the temporal attention is applied to automatically select the peak of expressions and aggregate the video-level representation. Our proposed CAN achieves the state-of-the-art performance on the three most widely used facial expression datasets: CK+ (99.03%), Oulu-CASIA (88.33%), and MMI (83.55%). Moreover, we conduct an extended experiment on a much more complex wild dataset AFEW and the experimental results further verify the generality of our attention mechanisms.

Long-range and short-range temporal modeling are two complementary and crucial aspects of video recognition. Most of the state-of-the-arts focus on short-range spatio-temporal modeling and then average multiple snippet-level predictions to yield the final video-level prediction. Thus, their video-level prediction does not consider spatio-temporal features of how video evolves along the temporal dimension. In this paper, we introduce a novel Dynamic Segment Aggregation (DSA) module to capture relationship among snippets. To be more specific, we attempt to generate a dynamic kernel for a convolutional operation to aggregate long-range temporal information among adjacent snippets adaptively. The DSA module is an efficient plug-and-play module and can be combined with the off-the-shelf clip-based models (i.e., TSM, I3D) to perform powerful long-range modeling with minimal overhead. The final video architecture, coined as DSANet. We conduct extensive experiments on several video recognition benchmarks (i.e., Mini-Kinetics-200, Kinetics-400, Something-Something V1 and ActivityNet) to show its superiority. Our proposed DSA module is shown to benefit various video recognition models significantly. For example, equipped with DSA modules, the top-1 accuracy of I3D ResNet-50 is improved from 74.9% to 78.2% on Kinetics-400. Codes are available at https://github.com/whwu95/DSANet., Comment: Accepted to ACMMM2021

Crowd counting has drawn much attention due to its importance in safety-critical surveillance systems. Especially, deep neural network (DNN) methods have significantly reduced estimation errors for crowd counting missions. Recent studies have demonstrated that DNNs are vulnerable to adversarial attacks, i.e., normal images with human-imperceptible perturbations could mislead DNNs to make false predictions. In this work, we propose a robust attack strategy called Adversarial Patch Attack with Momentum (APAM) to systematically evaluate the robustness of crowd counting models, where the attacker's goal is to create an adversarial perturbation that severely degrades their performances, thus leading to public safety accidents (e.g., stampede accidents). Especially, the proposed attack leverages the extreme-density background information of input images to generate robust adversarial patches via a series of transformations (e.g., interpolation, rotation, etc.). We observe that by perturbing less than 6\% of image pixels, our attacks severely degrade the performance of crowd counting systems, both digitally and physically. To better enhance the adversarial robustness of crowd counting models, we propose the first regression model-based Randomized Ablation (RA), which is more sufficient than Adversarial Training (ADT) (Mean Absolute Error of RA is 5 lower than ADT on clean samples and 30 lower than ADT on adversarial examples). Extensive experiments on five crowd counting models demonstrate the effectiveness and generality of the proposed method. The supplementary materials and certificate retrained models are available at \url{https://www.dropbox.com/s/hc4fdx133vht0qb/ACM_MM2021_Supp.pdf?dl=0}, Comment: Accepted by ACM Multimedia 2021

Recent deep networks have convincingly demonstrated high capability in crowd counting, which is a critical task attracting widespread attention due to its various industrial applications. Despite such progress, trained data-dependent models usually can not generalize well to unseen scenarios because of the inherent domain shift. To facilitate this issue, this paper proposes a novel adversarial scoring network (ASNet) to gradually bridge the gap across domains from coarse to fine granularity. In specific, at the coarse-grained stage, we design a dual-discriminator strategy to adapt source domain to be close to the targets from the perspectives of both global and local feature space via adversarial learning. The distributions between two domains can thus be aligned roughly. At the fine-grained stage, we explore the transferability of source characteristics by scoring how similar the source samples are to target ones from multiple levels based on generative probability derived from coarse stage. Guided by these hierarchical scores, the transferable source features are properly selected to enhance the knowledge transfer during the adaptation process. With the coarse-to-fine design, the generalization bottleneck induced from the domain discrepancy can be effectively alleviated. Three sets of migration experiments show that the proposed methods achieve state-of-the-art counting performance compared with major unsupervised methods., Accepted by ACMMM2021

Video-based person re-identification aims to retrieve video sequences of the same person in the multi-camera system. In this paper, we propose a k -reciprocal harmonious attention network (KHAN) to jointly learn discriminative spatiotemporal features and the similarity metrics. In KHAN, the harmonious attention module adaptively calibrates response at each spatial position and each channel by explicitly inspecting position-wise and channel-wise interactions over feature maps. Besides, the k -reciprocal attention module attends key features from all frame-level features with a discriminative feature selection algorithm; thus, useful temporal information within contextualized key features can be assimilated to produce more robust clip-level representation. Compared with commonly used local-context based approaches, the proposed KHAN captures long dependency of different spatial regions and visual patterns while incorporating informative context at each time-step in a non-parametric manner. The extensive experiments on three public benchmark datasets show that the performance of our proposed approach outperforms the state-of-the-art methods.

An essential yet challenging issue in crowd counting is the diverse background variations under complicated real-life environments, which makes attention based methods favorable in recent years. However, most existing methods only rely on first-order attention schemes (e.g. 2D position-wise attention), while ignoring the higher-order information within the congested scenes completely. In this paper, we propose a hybrid attention-aware network (HANet) with a high-order attention module (HAM) and an adaptive compensation loss (ACLoss) to tackle this problem. On the one hand, the HAM applies 3D attention to capture the subtle discriminative features around each people in the crowd. On the other hand, with the distributed supervision, the ACLoss exploits the prior knowledge from higher-level stages to guide the density map prediction at a lower level. The proposed HANet is then established with HAM and ACLoss working as different roles and promoting each other. Extensive experimental results show the superiority of our HANet against the state-of-the-arts on three challenging benchmarks.

Background Nonsuicidal self-injury (NSSI) is related to childhood abuse, family relationships, interpersonal relationships, personality, but the interaction between them is still unclear, and how they interact in major depressive disorder (MDD). Methods Collected data from 444 undergraduate degree students with MDD participated. Used the Eysenck Personality Questionnaire, Childhood Trauma Questionnaire, Interpersonal Relationship Comprehensive Diagnostic Scale, and Family Assessment Device to assess the patients' psychosocial factors. NSSI behavior was assessed through interviews. Use the Chi-square test, Wilcoxon rank-sum test, Kruskal-Wallis H-test, Distance Correlation, Structural Equation Mode for data analysis. Results Overall, 34.2% of patients with MDD had a history of NSSI. MDD patients with a history of NSSI had significant differences in psychoticism, neuroticism, emotional abuse, physical abuse, sexual abuse, emotional neglect, troubles in conversation, ability to make friends and family roles. Among these factors, psychoticism was most related to NSSI, and child abuse, interpersonal relationships and family roles played a variety of roles in mediating the relationship with NSSI. Conclusions Psychosocial factors affect the occurrence of NSSI through chain intermediary effects.

Male infertility is a complicated disease with causes generally split into 2 broad categories: genetic factors and environmental factors. The present study was designed to investigate the association between the methylation patterns of HI9 and SNRPN imprinting control region (ICR) and male infertility and to assess the gene-environment interactions between environmental factors and methylation patterns. A total of 205 DNA samples from 48 oligozoospermia (OZ), 52 asthenozoospermia (AZ), 55 teratozoospermia (TZ) patients, and 50 normozoospermia (NZ) men were analyzed. The mean methylation level of H/9-ICR in OZ (80.40% ± 12.74%) and AZ patients (81.17% ± 13.18%) was significantly lower than methylation in men with NZ (88.51 % ± 10.54%, P

Cross-domain sentiment classification aims to predict sentiment polarity on a target domain utilizing a classifier learned from a source domain. Most existing adversarial learning methods focus on aligning the global marginal distribution by fooling a domain discriminator, without taking category-specific decision boundaries into consideration, which can lead to the mismatch of category-level features. In this work, we propose an adversarial category alignment network (ACAN), which attempts to enhance category consistency between the source domain and the target domain. Specifically, we increase the discrepancy of two polarity classifiers to provide diverse views, locating ambiguous features near the decision boundaries. Then the generator learns to create better features away from the category boundaries by minimizing this discrepancy. Experimental results on benchmark datasets show that the proposed method can achieve state-of-the-art performance and produce more discriminative features.

Aim In the present study, we investigated whether the therapeutic dosages of Ginkgo biloba extract (EGb) and Aspirin (ASP) might synergistically suppress oxidative stress through regulating the expressions of LOX-1 and phosphorylated p38MAPK (p-p38MAPK) in human coronary artery endothelial cells (HCAECs) ex vivo . Methods HCAECs were stressed with activated platelets (2 × 10 8 /ml) and followed by ASP (1, 2 or 5 mmol/l), EGb (4, 40 or 400 μg/ml) and combinational (1 mmol/l ASP and 40 μg/ml EGb) treatments in three groups for 12 h. Superoxide anion in HCAECs was measured with H2DCF-DA probe. The expressions of LOX-1 and p-p38MAPK were examined by Western blot. Results After stimulation of activated platelets, intracellular superoxide anion was increased about 3-folds in HCAECs. Both ASP and EGb reduced superoxide anion in HCAECs in a dosage dependent manner. Combinational administration of ASP and EGb showed synergistic effect. By Western blot analysis, we were able to show that activated platelets markedly enhanced the expressions of LOX-1 and p-p38MAPK. Both ASP and EGb only inhibited LOX-1 expression in a concentration-dependent manner, but not p-p38MAPK. As expected, the combination of ASP and EGb markedly reduced not only the expression of LOX-1 but also the phosphorylation of p38MAPK. Conclusions Both EGb and ASP attenuate the oxidative stress of HCAECs stimulated by activated platelets ex vivo . It appears that the synergistic effect of EGb and ASP may correlate with the inhibition of ROS production, LOX-1 expression and p38MAPK phosphorylation.

Although many epidemiologic studies have investigated the FXIII-A Val34Leu polymorphism and their associations with myocardial infarction (MI), definite conclusions can’t be drawn. To clarify the effects of FXIII-A Val34Leu polymorphism on the risk of MI, a meta-analysis was performed. Related studies were identified from PubMed, Springer Link, Ovid, Chinese Wanfang Data Knowledge Service Platform, Chinese National Knowledge Infrastructure (CNKI), and Chinese Biology Medicine (CBM) till 10 August 2014. Pooled ORs and 95% CIs were used to assess the strength of the associations. A total of 12 studies including 3139 MI cases and 6343 healthy controls were involved in this meta-analysis. A significantly decreased MI risk was found (adjusted OR = 0.70, 95% CI 0.60-0.82, P < 0.00001). In the subgroup analysis by age, significantly decreased risks were found in the young population (OR = 0.70, 95% CI 0.54-0.91, P = 0.008) and old population (OR = 0.63, 95% CI 0.50-0.80, P = 0.0001). In the subgroup analysis by gender, significantly decreased risks were found in male (OR = 0.55, 95% CI 0.34-0.88, P = 0.01) and female (OR = 0.72, 95% CI 0.55-0.95, P = 0.02). When we limited the meta-analysis to studies that controlled for confounders such as age, sex, BMI, smoking, diabetes, hypertension, dyslipidemia, and fibrinogen, a significant association between FXIII-A Val34Leu polymorphism and MI risk remained. This meta-analysis provides the evidence that FXIII-A Val34Leu polymorphism may significant associated with the MI risk.