Search

Thallium exposure has been associated with female reproductive health, but little is known about its potential association with premature ovarian insufficiency (POI). In this study, a total of 169 patients with POI and 209 healthy women were recruited from Zhejiang province, China. Urinary thallium concentrations were significantly positively associated with the risk of POI [adjusted odds ratio (OR) = 1.63, 95% CI: 1.25-2.13, p < 0.001], geometric mean values of which were significantly higher in POI cases (0.213 μg/L, 0.302 μg/g for creatinine adjustment) than those of controls (0.153 μg/L, 0.233 μg/g for creatinine adjustment). Furthermore, the serum levels of follicle-stimulating hormone and luteinizing hormone were positively associated with urinary thallium concentrations, whereas anti-Mullerian hormone and estradiol were negatively correlated with thallium. To the best of our knowledge, this is the first study to provide evidence that thallium exposure at currently environmental levels is the potential risk factor for POI in women.

Objective The purpose of this survey is to evaluate the knowledge of Chinese pregnant women about fetal chromosomal aneuploidy and microdeletion and microduplication syndromes, and Non-invasive prenatal testing (NIPT). Study Design Written questionnaires were distributed to pregnant women who visited the fetal medicine clinic of the third affiliated hospital of Guangzhou medical university. A total of 330 questionnaires were given. Twenty-two questionnaires with incomplete information were excluded from further analysis. The remaining 308 cases were incorporated into the final analysis. Data were analyzed using IBM SPSS Statistics 26. Comparisons between categorical variables were tested by the use of crosstabs and χ 2 test. Results Among pregnant women, the recognition of Down syndrome was the highest (93.5 %), followed by maternal serum screening (74.0 %) and NIPT (69.2 %) for chromosomal aneuploidy. The awareness rates of chromosomal microdeletions and microduplications (18.2 %) and monogenic disorders (13.3 %) were the lowest. There were no significant differences in age, education, and conception way between pregnant women (P > 0.05). When asked the opinion on increasing the testing range of NIPT, more than 50 % of pregnant women chose to follow the provider's advice. Conclusion More than half of pregnant women lacked knowledge of screening methods for fetal chromosomal abnormalities. Less than 20 % of pregnant women knew microdeletion microduplication syndromes. To enhance the understanding of chromosomal abnormalities can ensure that women can actively choose tests rather than passively agreeing to their provider's recommendations so as to avoid missing the optimal prenatal screening time.

Cadmium, a toxic heavy metal that occurs in the environment in large quantities through human activities, has been shown to have adverse effects on female reproductive health. However, the association between cadmium exposure and primary ovarian insufficiency (POI), one of the most prevalent ovarian diseases in women, has not been examined yet. This case–control study involving 169 POI cases and 209 healthy controls was conducted in Zhejiang Province, China. The urinary concentrations of cadmium were determined by inductively coupled plasma mass spectrometry (ICP-MS). In addition, serum levels of reproductive hormones, including follicle-stimulating hormone (FSH), luteinizing hormone (LH), anti-Mullerian hormone (AMH) and estradiol, were measured. The median concentration of urinary cadmium in POI cases (0.43 μg/L, 0.58 μg/g for creatinine adjustment) was significantly higher than that of controls (0.29 μg/L, 0.43 μg/g for creatinine adjustment). The results of binary logistic regression models showed that the concentrations of urinary cadmium were positively significantly correlated with the odds ratio (ORs) of POI before the adjustment of confounders. After the adjustment, a significantly positive association was still present between the increased concentrations of cadmium and the ORs of POI (2.50, 95% CIs: 1.34–4.65 for the third tertile, p for trend = 0.001). The serum levels of FSH and LH were positively associated with urinary cadmium, while AMH and estradiol levels were inversely correlated. To the best of our knowledge, this is the first reported positive association of cadmium exposure with the risk of POI in women.

Polycyclic aromatic hydrocarbons (PAHs) are a typical class of persistent organic pollutants that is ubiquitous worldwide. Previous animal studies suggested that PAHs had adverse effects on female reproduction. However, the human data regarding relationship of PAHs exposure with women reproductive health, such as ovarian dysfunction, are scarce. In this case-control study, the associations of serum levels of PAHs with the risk of premature ovarian failure (POF) and reproductive hormones in Chinese women were investigated, with recruiting 157 POF patients and 217 healthy women. The serum levels of 12 types of PAHs, as well as reproductive hormones, including follicle-stimulating hormone, luteinizing hormone and anti-mullerian hormone, were determined. In the logistic regression models, most individual PAH congeners showed significantly positive correlations with the risk of POF (p 0.05), except for fluorine and pyrene. Benzo(a)pyrene (BaP), as the most carcinogenic PAH congener, was observed to be significantly positively associated with the risk of POF. After adjustment for age, body mass index, educational levels and household income, per one-unit increase in the log-transformed BaP concentration was significantly correlated with 2.191-fold increased risk of POF (OR = 2.191, 95%CI: 1.634-2.938, p 0.05). To the best of our knowledge, this is the first study to report an association between internal exposure levels of PAHs and the increased risk of POF in women.

Objective The L1 cell adhesion molecule (L1CAM) gene, encodes the L1 cell adhesion molecule, is involved in the central nervous system development. Its mutations result in L1 syndrome which is associated with brain malformation and nervous developmental delay. Case report We presented three fetuses with hydrocephalus and agenesis of the corpus callosum detected by ultrasound, followed by medical exome sequencing (MES) test with L1CAM mutations: two known missense mutation c.551G > A (p. R184Q) and c.1354G > A (p. G452R), and a novel frameshift mutation c.1322delG which causes the early termination of translation (p. G441Afs∗72). By utilizing multiple computational analysis, all the variants were scored to be likely pathogenic. Conclusion Combined use of ultrasound and MES to identify the molecular etiology of fetal anomalies may contribute to expanding our knowledge of the clinical phenotype of L1 syndrome observed in the south Chinese population.

Object reidentification with the goal of matching pedestrian or vehicle images captured from different camera viewpoints is of considerable significance to public security. Quadruple directional deep learning features (QD-DLFs) can comprehensively describe object images. However, the correlation among QD-DLFs is an unavoidable problem, since QD-DLFs are learned with quadruple independent directional deep networks (QIDDNs) driven with the same training data, and each network holds the same basic deep feature learning architecture (BDFLA). The correlation among QD-DLFs is harmful to the complementarity of QD-DLFs, restricting the object reidentification performance. For that, we propose joint quadruple decorrelation directional deep networks (JQD3Ns) to reduce the correlation among the learned QD-DLFs. In order to jointly train JQD3Ns, besides the softmax loss functions, a parameter correlation cost function is proposed to indirectly reduce the correlation among QD-DLFs by enlarging the dissimilarity among the parameters of JQD3Ns. Extensive experiments on three publicly available large-scale data sets demonstrate that the proposed JQD3Ns approach is superior to multiple state-of-the-art object reidentification methods.

Background Nonfunctioning pituitary adenoma (NFPA) and growth hormone pituitary adenoma (GHPA) are major subtypes of pituitary adenomas (PAs). The primary treatment is surgical resection. However, radical excision remains challenging, and few effective medical therapies are available. It is urgent to find novel targets for the treatment. Bromodomain-containing protein 4 (BRD4) is an epigenetic regulator that leads to aberrant transcriptional activation of oncogenes. Herein, we investigated the pathological role of BRD4 and evaluated the effectiveness of BRD4 inhibitors in the treatment of NFPA and GHPA. Methods The expression of BRD4 was detected in NFPA, GHPA, and normal pituitary tissues. The efficacies of BRD4 inhibitors were evaluated in GH3 and MMQ cell lines, patient-derived tumor cells, and in vivo mouse xenograft models of PA. Standard western blots, real-time PCR, and flow cytometry experiments were performed to investigate the effect of BRD4 inhibitors on cell cycle progression, apoptosis, and the expression patterns of downstream genes. Results Immunohistochemistry studies demonstrated the overexpression of BRD4 in NFPA and GHPA. In vitro and in vivo studies showed that treatment with the BRD4 inhibitor ZBC-260 significantly inhibited the proliferation of PA cells. Further mechanistic studies revealed that ZBC-260 could downregulate the expression of c-Myc, B-cell lymphoma 2 (Bcl2), and related genes, which are vital factors in pituitary tumorigenesis. Conclusion In this study, we determined the overexpression of BRD4 in NFPA and GHPA and assessed the effects of BRD4 inhibitors on PA cells in vitro and in vivo. Our findings suggest that BRD4 is a promising therapeutic target for NFPA and GHPA.

Homeostasis disturbance of trace elements has been linked to adverse reproductive consequences, including premature ovarian insufficiency (POI) in women, but limited evidence has been reported so far. This case-control study evaluated the associations between 5 common urinary trace elements [copper (Cu), manganese (Mn), Iron (Fe), Selenium (Se), and zinc (Zn)] and the odds for POI. Urinary concentrations of these 5 metals and serum levels of POI-related reproductive hormones of 169 cases and 209 healthy controls were measured. The urinary levels of Cu and Se in women with POI were significantly higher than those in the controls. The positive associations were observed between Cu levels and the odds of POI [for the medium tertile: odds ratio (OR) = 3.79, 95% CI: 1.98-7.27, p

Without appealing to exhaustive labeled data, self-supervised monocular depth estimation (MDE) plays a fundamental role in computer vision. Previous methods usually adopt a one-stage MDE network, which is insufficient to achieve high performance. In this paper, we dig deep into this task to propose an aggressive framework termed AggNet. The framework is based on a training-only progressive two-stage module to perform pseudo counter-surveillance as well as a simple yet effective dual-warp loss function between image pairs. In particular, we first propose a residual module, which follows the MDE network to learn a refined depth. The residual module takes both the initial depth generated from MDE and the initial color image as input to generate refined depth with residual depth learning. Then, the refined depth is leveraged to supervise the initial depth simultaneously during the training period. For inference, only the MDE network is retained to regress depth from a single image, which gains better performance without introducing extra computation. In addition to self-distillation loss, a simple yet effective dual-warp consistency loss is introduced to encourage the MDE network to keep depth consistency between stereo image pairs. Extensive experiments show that our AggNet achieves state-of-the-art performance on the KITTI and Make3D datasets.

Joint learning of scene parsing and depth estimation remains a challenging task due to the rivalry between the two tasks. In this paper, we revisit the mutual enhancement for joint semantic segmentation and depth estimation. Inspired by the observation that the competition and cooperation could be reflected in the feature frequency components of different tasks, we propose a Frequency Aware Feature Enhancement (FAFE) network that can effectively enhance the reciprocal relationship whereas avoiding the competition. In FAFE, a frequency disentanglement module is proposed to fetch the favorable frequency component sets for each task and resolve the discordance between the two tasks. For task cooperation, we introduce a re-calibration unit to aggregate features of the two tasks, so as to complement task information with each other. Accordingly, the learning of each task can be boosted by the complementary task appropriately. Besides, a novel local-aware consistency loss function is proposed to impose on the predicted segmentation and depth so as to strengthen the cooperation. With the FAFE network and new local-aware consistency loss encapsulated into the multi-task learning network, the proposed approach achieves superior performance over previous state-of-the-art methods. Extensive experiments and ablation studies on multi-task datasets demonstrate the effectiveness of our proposed approach.

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

In this paper, we propose a dimension apart network (DANet) for radar object detection task. A Dimension Apart Module (DAM) is first designed to be lightweight and capable of extracting temporal-spatial information from the RAMap sequences. To fully utilize the hierarchical features from the RAMaps, we propose a multi-scale U-Net style network architecture termed DANet. Extensive experiments demonstrate that our proposed DANet achieves superior performance on the radar detection task at much less computational cost, compared to previous pioneer works. In addition to the proposed novel network, we also utilize a vast amount of data augmentation techniques. To further improve the robustness of our model, we ensemble the predicted results from a bunch of lightweight DANet variants. Finally, we achieve 82.2% on average precision and 90% on average recall of object detection performance and rank at 1st place in the ROD2021 radar detection challenge. Our code is available at: \urlhttps://github.com/jb892/ROD2021_Radar_Detection_Challenge_Baidu.

The objective of this paper is to learn context- and depth-aware feature representation to solve the problem of monocular 3D object detection. We make following contributions: (i) rather than appealing to the complicated pseudo-LiDAR based approach, we propose a depth-conditioned dynamic message propagation (DDMP) network to effectively integrate the multi-scale depth information with the image context;(ii) this is achieved by first adaptively sampling context-aware nodes in the image context and then dynamically predicting hybrid depth-dependent filter weights and affinity matrices for propagating information; (iii) by augmenting a center-aware depth encoding (CDE) task, our method successfully alleviates the inaccurate depth prior; (iv) we thoroughly demonstrate the effectiveness of our proposed approach and show state-of-the-art results among the monocular-based approaches on the KITTI benchmark dataset. Particularly, we rank $1^{st}$ in the highly competitive KITTI monocular 3D object detection track on the submission day (November 16th, 2020). Code and models are released at \url{https://github.com/fudan-zvg/DDMP}, Comment: CVPR 2021. Code at https://github.com/fudan-zvg/DDMP

Although some persistent organic pollutants (POPs) may not be currently in production, many of these chemicals still exist ubiquitously in the environment and continue to represent significant environmental health concern. One of the important concerns of POPs exposure may lead to the adverse reproductive consequences including primary ovarian insufficiency (POI), but the limited evidence has been reported so far. Our case-control study was performed to explore the association between serum levels of selected POPs and risk for POI in a Chinese female population, including 157 cases and 217 healthy controls. Serum concentrations of polychlorinated biphenyls (PCBs), organochlorine pesticides (OCPs), polybrominated diphenyl ethers (PBDEs) and serum levels of reproductive hormones [follicle-stimulating hormone (FSH), luteinizing hormone (LH), anti-Mullerian hormone (AMH) and estradiol] were measured. The medians of dioxin-like PCBs (DL-PCBs) and p,p′-dichlorodiphenyltrichloroethane (p,p′-DDT) were significantly higher in case sera than in controls. Higher concentrations of DL-PCBs and p,p’-DDT were significantly associated with elevated risk of POI. A dose-response relationship was observed between the sum of DL-PCBs (Σ6 DL-PCBs) and the risk of POI (p for trend = 0.003), with odds ratios for the second and third tertiles, compared with the first, of 1.31- (95% CIs: 0.67–2.57) and 3.15-fold (95% CIs: 1.63–6.10), respectively. Similar results were observed for the sum of p,p′-DDT and its metabolites (Σ2 p,p′-DDTs). In control women, FSH levels were negatively associated with PCBs exposure, while LH had a positive association with OCPs. In patients with POI, exposure to PCBs was correlated with higher levels of LH, whereas exposure to OCPs was associated with lower levels of AMH. To the best of our knowledge, this study for the first provided evidence that exposure to PCBs and DDT could be the potential risk factors for POI in women. Keywords: Persistent organic pollutants, Polychlorinated biphenyls, Organochlorine pesticides, Primary ovarian insufficiency, Reproductive health, Women

Very deep convolutional neural networks (CNNs) have recently achieved great success in stereo matching. It is still highly desirable to learn a robust feature map to improve ill-posed regions, such as weakly textured regions, reflective surfaces, and repetitive patterns. Therefore, we propose an end-to-end multi-dimensional residual dense attention network (MRDA-Net) in this paper, focusing on more comprehensive pixel-wise feature extraction. Our proposed network consists of two parts: the 2D residual dense attention net for feature extraction and the 3D convolutional attention net for matching. Our designed 2D residual dense attention net uses a dense network structure to fully exploit the hierarchical features from preceding convolutional layers and uses residual network structure to fuse low-level structure information and high-level semantic information. The 2D attention module of the net aims to adaptively recalibrate channel-wise features to be more concerned about informative features. Our proposed 3D convolutional attention net further expands attention mechanism for matching. The stacked hourglass module of the net is exploited to extract multi-scale context information as well as geometry information. The novel 3D attention module of the net aggregates hierarchical sub-cost volumes adaptively instead of manually and then achieves a comprehensive recalibrated cost volume for more correct disparity computation. The experiments demonstrate that our approach achieves the state-of-the-art accuracy on Scene Flow dataset and KITTI 2012 and KITTI 2015 Stereo datasets.

This paper presents a semantic-guided interpolation scheme (SemFlow) to handle motion boundaries and occlusions in large displacement optical flow. The basic idea is to segment images into superpixels and estimate their homographies for interpolation. In order to ensure each superpixel can be approximated as a plane, a semantic-guided refinement method is introduced. Moreover, we put forward a homography estimation model weighted by the distance between each superpixel and its K -nearest neighbors. Our newly-proposed distance metric combines the texture and semantic information to find proper neighbors. Our homography model performs better than the original affine model, since it accords with the real world projection relationship. The experiments on KITTI dataset demonstrate that SemFlow outperforms other state-of-the-art methods, especially in solving the problem of large scale motions and occlusions.

Despite the tremendous progress made in learning-based depth prediction, most methods rely heavily on large amounts of dense ground-truth depth data for training. To solve the tradeoff between the labeling cost and precision, we propose a novel weakly supervised approach, namely, the Guided-Net, by incorporating robust ground control points for guidance. By exploiting the guidance from ground control points, disparity edge gradients, and image appearance constraints, our improved network with deformable convolutional layers is empowered to learn in a more efficient way. The experiments on the KITTI, Cityscapes, and Make3D datasets demonstrate that the proposed method yields a performance superior to that of the existing weakly supervised approaches and achieves results comparable to those of the semisupervised and supervised frameworks.

Background To assess the indications and complications of late amniocentesis and the advanced genetic test results in a tertiary university fetal medical medicine unit. Methods In this retrospective study, women that underwent amniocentesis at 24+ 0 to 39+ 4 weeks, between January 2014 and December 2019, were recruited. Indications, complications, genetic test results, and pregnancy outcomes were reported for each pregnancy and compared with those who underwent the traditional amniocentesis at 16+ 0 to 23+ 6 weeks (control group). Information was retrieved from patient medical records, checked by research staff, and analyzed. Results Of the 1287 women (1321 fetuses) included in the late amniocentesis group, late detected sonographic abnormalities (85.5%) were the most common indication. The overall incidence of preterm birth and intrauterine demise after amniocentesis were 2.5 and 1.3%, respectively. Sixty-nine fetuses with aneuploidy (5.3%) and seventy-two fetuses with pathogenic copy number variations (5.5%) were identified by chromosomal microarray analysis. The maximal diagnostic yield (70%) was in the subgroup of fetuses with the abnormal diagnostic test results, followed by abnormal NIPT results (35.7%) and multiple abnormalities (23.8%). And 35.4% of the pregnancies were finally terminated. Conclusions Due to the high detection rates of advanced genetic technologies and the safety of the invasive procedure (3.9% vs 4.0%), it is reasonable to recommend late amniocentesis as an effective and reliable method to detect late-onset fetal abnormalities. However, chromosomal microarray and whole-exome sequencing may result in uncertain results like variants of uncertain significance. Comprehensive genetic counseling is necessary.

Objective: To investigate the effects of microRNA-137 (MIR137) polymorphisms (rs1198588 and rs2660304) on the risk of schizophrenia in a Han Chinese population.Methods: Schizophrenia was diagnosed according to the DSM-5. Clinical symptoms and cognitive functions were assessed with the Positive and Negative Symptom Scale (PANSS) and Brief Assessment of Cognition in Schizophrenia (BACS), respectively. The polymorphisms were genotyped by improved multiplex ligation detection reaction (iMLDR) technology in 1,116 patients with schizophrenia and 1,039 healthy controls.Results: Significant associations were found between schizophrenia and MIR137 in the distributions of genotypes (p = 0.037 for rs1198588; p = 0.037 for rs2660304, FDR corrected) and alleles (p = 0.043 for rs1198588; p = 0.043 for rs2660304, FDR corrected) of two SNPs. When the population was stratified by sex, we found female-specific associations between MIR137 and schizophrenia in terms of genotype and allele distributions of rs1198588 (χ2 = 4.41, p = 0.036 and χ2 = 4.86, p = 0.029, respectively, FDR corrected) and rs2660304 (χ2 = 4.74, p=0.036 and χ2 = 4.80, p = 0.029, respectively, FDR corrected). Analysis of the MIR137 haplotype rs1198588-rs2660304 showed a significant association with schizophrenia in haplotype T-T [χ2 = 4.60, p = 0.032, OR = 1.32, 95% CI (1.02–1.70)]. Then, significant female-specific associations were found with the haplotypes T-T and G-A [χ2 = 4.92, p = 0.027, OR = 1.62, 95% CI (1.05–2.50); χ2 = 4.42, p = 0.035, OR = 0.62, 95% CI (0.39–0.97), respectively]. When the TT genotype of rs1198588 was compared to the GT+GG genotype, a clinical characteristics analysis also showed a female-specific association in category instances (t = 2.76, p = 0.042, FDR corrected).Conclusion: The polymorphisms within the MIR137 gene are associated with susceptibility to schizophrenia, and a female-specific association of MIR137 with schizophrenia was reported in a Han Chinese population.

Neural networks have recently achieved impressive success in semantic and instance segmentation on 2D images. However, their capabilities have not been fully explored to address semantic instance segmentation on unstructured 3D point cloud data. Digging into the regional feature representation to boost point cloud comprehension, we propose a region-feature-enhanced structure consisting of adaptive regional feature complementary (ARFC) module and affinity-based regional relational reasoning (AR3) module. The ARFC module aims to complement low-level features of sparse regions adaptively. The AR3 module emphasizes on mining the potential reasoning relationships between high-level features based on affinity. Both the ARFC and AR3 modules are plug-and-play. Besides, a novel dual spatial-aware discriminative loss is proposed to improve the discrimination of instance embedding. Our proposal-free point cloud instance segmentation network (RegionNet) equipped with the region-feature-enhanced structure and dual spatial-aware discriminative loss achieves state-of-the-art performance on S3DIS dataset and ScanNet-v2 dataset.

Object detection from 3D point clouds remains a challenging task, though recent studies pushed the envelope with the deep learning techniques. Owing to the severe spatial occlusion and inherent variance of point density with the distance to sensors, appearance of a same object varies a lot in point cloud data. Designing robust feature representation against such appearance changes is hence the key issue in a 3D object detection method. In this paper, we innovatively propose a domain adaptation like approach to enhance the robustness of the feature representation. More specifically, we bridge the gap between the perceptual domain where the feature comes from a real scene and the conceptual domain where the feature is extracted from an augmented scene consisting of non-occlusion point cloud rich of detailed information. This domain adaptation approach mimics the functionality of the human brain when proceeding object perception. Extensive experiments demonstrate that our simple yet effective approach fundamentally boosts the performance of 3D point cloud object detection and achieves the state-of-the-art results., Comment: 8 pages, 5 figures, CVPR 2020

With the demands of intelligent traffic, vehicle counting has become a vital problem, which can be used to mitigate traffic congestion and elevate the efficiency of the traffic light. Traditional vehicle counting problems focus on counting vehicles in a single frame or consecutive frames. Nevertheless, they are not expected to count vehicles by movements of interest (MOI), which can be pre-defined by all possible states of vehicles, combining different lanes and directions. In this paper, we mainly focus on movement-specific vehicle counting problem. A detection-tracking-counting (DTC) framework is applied, which detects and tracks objects in the region of interest (ROI), then counts those tracked trajectories by movements. To be specific, we propose the detection augmentation method and the Mahalanobis distance smoothness method to improve the multi-object tracking performance. For vehicle counting, a shape-based movement assignment method is carefully designed to categorize each trajectory by movements. Experiments are conducted on both the AICity 2020 Track-1 Dataset and the Vehicle-Track Dataset, which is built in this paper. Experimental results show the effectiveness and efficiency of our method.

Objective To explore the clinical utility of detecting chromosome copy number variants (CNVs) in the fetus by noninvasive prenatal testing (NIPT) using the low-pass whole-genome sequencing. Methods Eight hundred and seventy-three singleton pregnancies with chromosomal microarray analysis (CMA) available between January 2017 to December 2019 and stored enough plasma sample for NIPT testing were included in this study. The CMA results show that forty-eight pregnancies with CNVs and eight hundred and twenty-five pregnancies are normal. Each pregnancy's plasma sample was blindly tested with NIPT at a depth of 0.51-1.19x for CNVs detection. The performance of the NIPT method for CNVs detection compared with the CMA method is evaluated. Results A total of fifty-two CNVs ranging from 0.1–47.3 Mb identified in forty-eight samples were identified by NIPT, of which thirty-four CNVs were consistent with CMA results. Additionally, eighteen CNVs were missed by NIPT. The overall sensitivity and specificity for the detection of CNVs were 65.38% (95% CI: 51.76%-76.89%) and 97.45% (95% CI: 96.12%-98.35%), respectively. However, for the detection of CNVs larger than 2 Mb and CNVs less than 2Mb, the sensitivities were 81.58% (95% CI: 66.27%-91.09%) and 21.43% (95% CI: 6.84%-48.32%), respectively. Conclusion Our study demonstrated that the NIPT might be an alternative method for screening CNVs comparable with other studies. However, CNVs less than 2Mb in length shows poor sensitivity by NIPT. Noninvasive CNVs detection based on the NIPT method still needs more clinical validation studies and technical improvement to achieve clinically acceptable accuracy.

A few epidemiological investigations and animal studies have demonstrated that bisphenol A (BPA) may affect female reproductive health. However, no epidemiologic study has investigated the relationship between BPA exposure and the risk of premature ovarian insufficiency (POI). In this case–control study, urinary concentrations of BPA and serum levels of reproductive hormone were measured. Associations between BPA concentrations and the risk of POI and POI-related hormone levels were estimated. Among BPA quartiles, no obvious association was found between BPA levels and the risk of POI (p = 0.603). Although the adjusted odds ratio (OR) of POI was slightly increased for participants in the highest BPA concentration quartile, the association was not statistically significant (OR = 1.282, 95% confidence interval [CI] 0.615–2.049 for the highest vs. lowest quartile, p = 0.508). Although follicle stimulating hormone (FSH) and anti-Mullerian hormone (AMH) levels showed no tendency of an association with BPA (p = 0.941 and p = 0.876 for FSH and AMH, respectively), the highest quartile of luteinizing hormone was significantly positively associated with BPA levels (OR = 1.333, 95% CI 0.986–1.803, p = 0.042). The urinary concentrations of BPA determined in this study were consistent with the range of exposure currently observed in Chinese women. However, BPA exposure at a relatively low level is not associated with POI in Chinese women. Further epidemiological studies are needed to confirm our findings.

Monocular 3D object detection is a challenging task due to unreliable depth, resulting in a distinct performance gap between monocular and LiDAR-based approaches. In this paper, we propose a novel domain adaptation based monocular 3D object detection framework named DA-3Ddet, which adapts the feature from unsound image-based pseudo-LiDAR domain to the accurate real LiDAR domain for performance boosting. In order to solve the overlooked problem of inconsistency between the foreground mask of pseudo and real LiDAR caused by inaccurately estimated depth, we also introduce a context-aware foreground segmentation module which helps to involve relevant points for foreground masking. Extensive experiments on KITTI dataset demonstrate that our simple yet effective framework outperforms other state-of-the-arts by a large margin.

3D object detection is an essential task in autonomous driving and robotics. Though great progress has been made, challenges remain in estimating 3D pose for distant and occluded objects. In this paper, we present a novel framework named ZoomNet for stereo imagery-based 3D detection. The pipeline of ZoomNet begins with an ordinary 2D object detection model which is used to obtain pairs of left-right bounding boxes. To further exploit the abundant texture cues in RGB images for more accurate disparity estimation, we introduce a conceptually straight-forward module -- adaptive zooming, which simultaneously resizes 2D instance bounding boxes to a unified resolution and adjusts the camera intrinsic parameters accordingly. In this way, we are able to estimate higher-quality disparity maps from the resized box images then construct dense point clouds for both nearby and distant objects. Moreover, we introduce to learn part locations as complementary features to improve the resistance against occlusion and put forward the 3D fitting score to better estimate the 3D detection quality. Extensive experiments on the popular KITTI 3D detection dataset indicate ZoomNet surpasses all previous state-of-the-art methods by large margins (improved by 9.4% on APbv (IoU=0.7) over pseudo-LiDAR). Ablation study also demonstrates that our adaptive zooming strategy brings an improvement of over 10% on AP3d (IoU=0.7). In addition, since the official KITTI benchmark lacks fine-grained annotations like pixel-wise part locations, we also present our KFG dataset by augmenting KITTI with detailed instance-wise annotations including pixel-wise part location, pixel-wise disparity, etc.. Both the KFG dataset and our codes will be publicly available at https://github.com/detectRecog/ZoomNet., Comment: Accpeted by AAAI 2020 as Oral presentation; The github page will be updated in March,2020

Accurate stereo matching is still challenging in case of weakly textured areas, discontinuities, and occlusions. Besides, occlusion recovery is often regarded as a subordinate problem and simply handled. To obtain dense high-accuracy depth maps, this letter proposes an efficient multistep disparity refinement framework with occlusion handling. The framework is implemented by classifying the outliers into leftmost occlusions, nonborder occlusions, as well as mismatches, and employing different strategies to recover them. To recover occlusions, a filling order is specially introduced to avoid error propagation and surface decision based on local image content is performed when more than one background surface exists. The evaluations on Middlebury datasets and comparisons with other refinement algorithms show the superiority and robustness of our method.

Stereo matching is a challenging problem with respect to weak texture, discontinuities, illumination difference and occlusions. Therefore, a deep learning framework is presented in this paper, which focuses on the first and last stage of typical stereo methods: the matching cost computation and the disparity refinement. For matching cost computation, two patch-based network architectures are exploited to allow the trade-off between speed and accuracy, both of which leverage multi-size and multi-layer pooling unit with no strides to learn cross-scale feature representations. For disparity refinement, unlike traditional handcrafted refinement algorithms, we incorporate the initial optimal and sub-optimal disparity maps before outlier detection. Furthermore, diverse base learners are encouraged to focus on specific replacement tasks, corresponding to the smooth regions and details. Experiments on different datasets demonstrate the effectiveness of our approach, which is able to obtain sub-pixel accuracy and restore occlusions to a great extent. Specifically, our accurate framework attains near-peak accuracy both in non-occluded and occluded region and our fast framework achieves competitive performance against the fast algorithms on Middlebury benchmark.

Despite the great success achieved by supervised fully convolutional models in semantic segmentation, training the models requires a large amount of labor-intensive work to generate pixel-level annotations. Recent works exploit synthetic data to train the model for semantic segmentation, but the domain adaptation between real and synthetic images remains a challenging problem. In this work, we propose a Separated Semantic Feature based domain adaptation network, named SSF-DAN, for semantic segmentation. First, a Semantic-wise Separable Discriminator (SS-D) is designed to independently adapt semantic features across the target and source domains, which addresses the inconsistent adaptation issue in the class-wise adversarial learning. In SS-D, a progressive confidence strategy is included to achieve a more reliable separation. Then, an efficient Class-wise Adversarial loss Reweighting module (CA-R) is introduced to balance the class-wise adversarial learning process, which leads the generator to focus more on poorly adapted classes. The presented framework demonstrates robust performance, superior to state-of-the-art methods on benchmark datasets.

The dataset presented in this article supports “Selected persistent organic pollutants associated with the risk of primary ovarian insufficiency in women” (Pan et al., 2019). The supplementary data were as follows: (1) Detailed information regarding pretreatment methods, instrumental analysis and methods validation of quantification of serum concentrations of persistent organic pollutants (POPs). (2) The total dioxin equivalents (TEQs) levels of dioxin-like PCBs (DL-PCBs) in primary ovarian insufficiency (POI) cases and controls, as well as the association of TEQ levels with the risk of POI. (3) The results of principal components analyses (PCA) about 20 POPs that were detected in >40% samples. Keywords: Persistent organic pollutants, Polychlorinated biphenyls, Organochlorine pesticides, Primary ovarian insufficiency, Pretreatment methods

Arsenic, a well-known toxic metalloid, is ubiquitously existed in environment. Arsenic exposure has been associated with female reproductive health. However, a potential association between arsenic exposure and primary ovarian insufficiency (POI) in women has not been recognized yet. In this case-control study, a total of 169 POI cases and 209 healthy controls were recruited to determine urinary concentrations of arsenic and serum levels of reproductive hormones (follicle-stimulating hormone (FSH), luteinizing hormone (LH), anti-Mullerian hormone (AMH) and estradiol). The median concentration of urinary arsenic in cases (21.5 μg/L, 28.0 μg/g for creatinine adjustment) was significantly higher than that of controls (13.8 μg/L, 19.3 μg/g for creatinine adjustment). Urinary arsenic concentrations were significantly positively associated with the risk of POI (adjusted odds ratio (OR) = 2.66, 95% CI: 1.43–4.95 for the highest vs lowest tertile of arsenic, p = 0.002; p for trend = 0.004). We also assessed the associations between arsenic exposure and reproductive hormones that are important for ovarian functions. FSH and LH levels were positively associated with urinary arsenic, whereas AMH and estradiol levels were negatively correlated with this element. This study provided evidence that arsenic exposure could be the potential risk factor for POI in women.

Owing to the imbalance of observed data and user’s preference, it is necessary and meaningful to think over the multilevel characteristics of recommendation information (RI) during the recommendation process. At the same time, to better summarize user’s preference and enhance user’s beliefs, the interpretability of recommendation results also become more and more important in recommender system (RS). In view of the multilevel characteristics of RI and the interpretability of recommendation results, this paper proposes a novel interpretable sequential three-way recommendation strategy, namely, CTR-based cost-sensitive sequential three-way recommendation (CTR-CS3WR). First, in order to construct the interpretable granular features and multilevel information, we introduce collaborative topic regression (CTR) and design three novel granulation methods: PMF-based, LDA-based and CTR-based granulation method. Then, with the consideration of decision cost and time cost, a sequential three-way recommendation strategy is proposed to realize the multilevel recommendation. Finally, extensive experiments on two CiteUlike datasets verify the effectiveness of our proposed granulation methods and recommendation strategy.

The human brain can effortlessly recognize and localize objects, whereas current 3D object detection methods based on LiDAR point clouds still report inferior performance for detecting occluded and distant objects: the point cloud appearance varies greatly due to occlusion, and has inherent variance in point densities along the distance to sensors. Therefore, designing feature representations robust to such point clouds is critical. Inspired by human associative recognition, we propose a novel 3D detection framework that associates intact features for objects via domain adaptation. We bridge the gap between the perceptual domain, where features are derived from real scenes with sub-optimal representations, and the conceptual domain, where features are extracted from augmented scenes that consist of non-occlusion objects with rich detailed information. A feasible method is investigated to construct conceptual scenes without external datasets. We further introduce an attention-based re-weighting module that adaptively strengthens the feature adaptation of more informative regions. The network's feature enhancement ability is exploited without introducing extra cost during inference, which is plug-and-play in various 3D detection frameworks. We achieve new state-of-the-art performance on the KITTI 3D detection benchmark in both accuracy and speed. Experiments on nuScenes and Waymo datasets also validate the versatility of our method., Comment: 12 pages

Pyrethroids, a class of ubiquitous insecticides, have been recognized as endocrine-disrupting chemicals (EDCs). A lot of studies have implied the endocrine-disrupting effects of pyrethroids on the hypothalamic-pituitary-gonadal (HPG) axis. However, there are few review articles regarding the effects of pyrethroids on the HPG axis of mammal and human, especially new research progress made in this area. The present review sums up the effects of pyrethroids on the HPG axis-related reproductive outcomes, including epidemiological investigations based on human biomonitoring, animal studies and in vitro tests. Mechanisms have described that the endocrine-disrupting effects of pyrethroids on mammal can be mediated via the interaction with steroid receptors, the direct action on ion channels and signaling molecules. Finally, we summarize the current research gaps and suggest future directions in this topic.

Pyrethroids are a class of widely used insecticides. Female animal studies suggested that pyrethroid exposure impaired ovarian function, which resulted in similar symptoms of primary ovarian insufficiency (POI). However, it is still unknown whether this association applies to women. In this case-control study, a total of 172 POI patients and 247 control women were recruited in Zhejiang, China. The urinary concentrations of metabolites of pyrethroids, 3-phenoxybenzoic acid (3-PBA) and 4-fluoro-3-phenoxybenzoic acid (4-F-3-PBA), as well as the serum concentrations of follicle-stimulating hormone (FSH), luteinizing hormone (LH) and anti-Mullerian hormone (AMH) were determined. The associations of pyrethroid metabolites with POI and POI-related hormones were accessed using unconditional logistic regression. Higher urinary levels of 3-PBA were significantly associated with increased risk of POI [adjusted odds ratio (OR) = 2.344, 95% CI: 1.193-4.607 for the highest vs lowest quartile of 3-PBA, p = 0.013]. Stratified analyses showed that each log increase in urinary 3-PBA concentration was significantly associated with an induction in odds of 51.0% being in the highest quartile of FSH and 28.6% being in the highest quartile of LH levels, whereas a 25.9% reduction in odds of being in the highest quartile of AMH levels (All p for trend0.05). To our knowledge, this is the first case-control study to report an association of pyrethroid exposure with increased risk of POI in women.

Diesel exhaust (DE) is one of the main sources of urban air pollution. An increasing number of evidence showed the association of air pollution with cardiovascular diseases. Pulmonary arterial hypertension (PAH) is one of the most disastrous vascular diseases, which results in right ventricular failure and death. However, the relationship of DE inhalation exposure with PAH is still unknown. In this study, male adult mice were exposed by inhalation to filtered ambient air (negative control), 10% O2 hypoxia (PAH-phenotype positive control), 350 μg/m3 particulate matter whole DE, or the combination of DE and hypoxic condition. DE inhalation induced PAH-phenotype accompanied with increased right ventricular systolic pressure (RVSP), right ventricle hypertrophy and pulmonary arterial thickening in a mouse model. DE exposure induced the proliferation of vascular smooth muscle cells (VSMCs) and apoptosis of endothelial cells in pulmonary artery. DE inhalation exposure induced an accumulation of CD45+ lymphocytes and CD68+ macrophages surrounding and infiltrating pulmonary arteriole. The levels of pro-inflammatory cytokines tumor necrosis factor (TNF-α), interleukin-6 (IL-6) and IL-13 produced by T helper 17 (Th17) and Th2 cells were markedly elevated in lung tissues of mice after DE inhalation exposure. Our findings suggest DE exposure induces PAH by activating Th17-skewed and Th2-droved responses, stimulating VSMCs proliferation and inducing endothelial cell apoptosis by the production of multifunctional pro-inflammatory cytokines, especially IL-6 and TNF-α. Considering the adverse impact of air pollution on health care, it is imperative to understand air pollution-induced susceptibility of progressive cardiopulmonary disease, such as PAH, and also elucidate critical mechanistic pathways which mediate pulmonary artery vascular remodeling and may serve as targets for preventive measures.

Pyrethroids, a class of ubiquitous insecticides, have been considered as endocrine-disrupting chemicals (EDCs). Female animal studies suggested that early-life pyrethroids exposure might delay puberty onset. However, it remains unclear whether this association applies to human populations. A total of 305 girls at the ages of 9–15 years old were recruited in Hangzhou, China in this study. The concentration of the common metabolite of pyrethroids, 3-phenoxybenzoic acid (3-PBA), was analyzed in urine samples to reflect the exposure level of pyrethroids. The associations of 3-PBA with pubertal stages were evaluated using a multinomial logistic regression model. The geometric mean level of 3-PBA was 1.11 μg/L (1.42 μg/g for creatinine-adjusted concentration). There was a significant 45% reduction in odds of being in breast stage 3 (B3) per one-unit increase in the log-transformed 3-PBA levels [OR = 0.55 (95%CI: 0.31–0.98), p = 0.042]. A similar negative association was found between urinary 3-PBA levels with later onset by pubic hair stage 2 (P2) [OR = 0.56 (95%CI: 0.36–0.90), p = 0.015]. Similar negative association was also observed between urinary 3-PBA levels and pubertal onset indicated by menarche [OR = 0.51 (95%CI: 0.28–0.93), p = 0.029]. For the first time to our knowledge, this work reveals that pyrethroids exposure may increase the risk of delayed pubertal onset in girls.

On October 5, 191 1, the Qing court announced the newly composed national anthem, Gongjiriou l^feiS [Strengthening our hold on the golden cup]. Within six days, the revolution broke out and within a few months the dynasty had ceased to exist. The expression wangguo zhiyin t ffl/^W) 'sound of a doomed state', comes from the yueji ^|B section of the Book of Rites, devoted to music. It might be translated more freely 'the death knell of a dynasty. It became a common expression throughout Chinese history, and was used to refer to the decadent style of music that comes into vogue towards the end of a dynasty. In the case of the Qing national anthem, the term wangguo zhiyin can be taken literally, as we just saw. This article concentrates on the criticism of the anthem's musical score made by a prominent court musician, Cao Xinquan W;L^« Cao was thoroughly familiar with the traditional Confucian view of music based upon principles expounded in the Book of Rites. He argued that the anthem ignored those principles and hastened the demise of the dynasty. His suggested revision of the anthem exemplifies the concept that the primary function of music is moral education, not entertainment.

Sleep disturbance is significantly associated with overall morbidity, mortality, and quality of life in end-stage renal disease patients. Although it was reported that the prevalence of poor sleep quality was 41-83% in hemodialysis patients, sleep disturbance was not well understood in peritoneal dialysis (PD) patients, especially in Asian PD patients. The present study attempted to assess the prevalence and related risk factors of sleep disturbance among continuous ambulatory peritoneal dialysis (CAPD) patients in Guangzhou, southern China.Pittsburgh Sleep Quality Index (PSQI) was used to assess the sleep quality in eligible 212 CAPD patients (53.3% men), average age 49.9 ± 16.7 years.The average global PSQI score of the CAPD patients was 9.5 ± 5.2, with 171 (80.4%) poor sleepers (PSQI scoresor = 5). The PSQI score was negatively correlated with serum albumin (r = -0.21, P = 0.003) subjective global assessment (SGA) score (r = -0.27, P = 0.001) and positively correlated with age (r = 0.34, P0.001), high-sensitivity C-reactive protein (r = 0.20, P = 0.008), level of calcium × phosphate product (r = 0.19, P = 0.009), cardiovascular disease (r = 0.17, P = 0.019), Charlson comorbidity index score (r = 0.21, P = 0.004), malnutrition-inflammation score (MIS) (r = 0.31, P0.001), and duration on CAPD (r = 0.20, P = 0.005). In multiple analysis, old age, high calcium × phosphate product, low SGA score (β = -0.23, P = 0.042), and high MIS (β = 0.30, P = 0.007) were independent predictors of sleep disturbance in CAPD patients.We clearly demonstrated the novel relationship of malnutrition and calcium × phosphate product with poor sleep quality in CAPD patients, which may be a part of the explanation for the strong links between sleep quality and overall morbidity and mortality.