Discriminative Suprasphere Embedding for Fine-Grained Visual Categorization

Despite the great success of the existing work in fine-grained visual categorization (FGVC), there are still several unsolved challenges, e.g., poor interpretation and vagueness contribution. To circumvent this drawback, motivated by the hypersphere embedding method, we propose a discriminative suprasphere embedding (DSE) framework, which can provide intuitive geometric interpretation and effectively extract discriminative features. Specifically, DSE consists of three modules. The first module is a suprasphere embedding (SE) block, which learns discriminative information by emphasizing weight and phase. The second module is a phase activation map (PAM) used to analyze the contribution of local descriptors to the suprasphere feature representation, which uniformly highlights the object region and exhibits remarkable object localization capability. The last module is a class contribution map (CCM), which quantitatively analyzes the network classification decision and provides insight into the domain knowledge about classified objects. Comprehensive experiments on three benchmark datasets demonstrate the effectiveness of our proposed method in comparison with state-of-the-art methods.

GNDAN: Graph Navigated Dual Attention Network for Zero-Shot Learning

Zero-shot learning (ZSL) tackles the unseen class recognition problem by transferring semantic knowledge from seen classes to unseen ones. Typically, to guarantee desirable knowledge transfer, a direct embedding is adopted for associating the visual and semantic domains in ZSL. However, most existing ZSL methods focus on learning the embedding from implicit global features or image regions to the semantic space. Thus, they fail to: 1) exploit the appearance relationship priors between various local regions in a single image, which corresponds to the semantic information and 2) learn cooperative global and local features jointly for discriminative feature representations. In this article, we propose the novel graph navigated dual attention network (GNDAN) for ZSL to address these drawbacks. GNDAN employs a region-guided attention network (RAN) and a region-guided graph attention network (RGAT) to jointly learn a discriminative local embedding and incorporate global context for exploiting explicit global embeddings under the guidance of a graph. Specifically, RAN uses soft spatial attention to discover discriminative regions for generating local embeddings. Meanwhile, RGAT employs an attribute-based attention to obtain attribute-based region features, where each attribute focuses on the most relevant image regions. Motivated by the graph neural network (GNN), which is beneficial for structural relationship representations, RGAT further leverages a graph attention network to exploit the relationships between the attribute-based region features for explicit global embedding representations. Based on the self-calibration mechanism, the joint visual embedding learned is matched with the semantic embedding to form the final prediction. Extensive experiments on three benchmark datasets demonstrate that the proposed GNDAN achieves superior performances to the state-of-the-art methods. Our code and trained models are available at https://github.com/shiming-chen/GNDAN.

Heralded Three-Photon Entanglement from a Single-Photon Source on a Photonic Chip

In the quest to build general-purpose photonic quantum computers, fusion-based quantum computation has risen to prominence as a promising strategy. This model allows a ballistic construction of large cluster states which are universal for quantum computation, in a scalable and loss-tolerant way without feed forward, by fusing many small n-photon entangled resource states. However, a key obstacle to this architecture lies in efficiently generating the required essential resource states on photonic chips. One such critical seed state that has not yet been achieved is the heralded three-photon Greenberger-Horne-Zeilinger (3-GHZ) state. Here, we address this elementary resource gap, by reporting the first experimental realization of a heralded 3-GHZ state. Our implementation employs a low-loss and fully programmable photonic chip that manipulates six indistinguishable single photons of wavelengths in the telecommunication regime. Conditional on the heralding detection, we obtain the desired 3-GHZ state with a fidelity 0.573±0.024. Our Letter marks an important step for the future fault-tolerant photonic quantum computing, leading to the acceleration of building a large-scale optical quantum computer.

Multilevel Contrastive Graph Masked Autoencoders for Unsupervised Graph-Structure Learning

Unsupervised graph-structure learning (GSL) which aims to learn an effective graph structure applied to arbitrary downstream tasks by data itself without any labels' guidance, has recently received increasing attention in various real applications. Although several existing unsupervised GSL has achieved superior performance in different graph analytical tasks, how to utilize the popular graph masked autoencoder to sufficiently acquire effective supervision information from the data itself for improving the effectiveness of learned graph structure has been not effectively explored so far. To tackle the above issue, we present a multilevel contrastive graph masked autoencoder (MCGMAE) for unsupervised GSL. Specifically, we first introduce a graph masked autoencoder with the dual feature masking strategy to reconstruct the same input graph-structured data under the original structure generated by the data itself and learned graph-structure scenarios, respectively. And then, the inter-and intra-class contrastive loss is introduced to maximize the mutual information in feature and graph-structure reconstruction levels simultaneously. More importantly, the above inter-and intra-class contrastive loss is also applied to the graph encoder module for further strengthening their agreement at the feature-encoder level. In comparison to the existing unsupervised GSL, our proposed MCGMAE can effectively improve the training robustness of the unsupervised GSL via different-level supervision information from the data itself. Extensive experiments on three graph analytical tasks and eight datasets validate the effectiveness of the proposed MCGMAE.

TransZero++: Cross Attribute-Guided Transformer for Zero-Shot Learning

Zero-shot learning (ZSL) tackles the novel class recognition problem by transferring semantic knowledge from seen classes to unseen ones. Semantic knowledge is typically represented by attribute descriptions shared between different classes, which act as strong priors for localizing object attributes that represent discriminative region features, enabling significant and sufficient visual-semantic interaction for advancing ZSL. Existing attention-based models have struggled to learn inferior region features in a single image by solely using unidirectional attention, which ignore the transferable and discriminative attribute localization of visual features for representing the key semantic knowledge for effective knowledge transfer in ZSL. In this paper, we propose a cross attribute-guided Transformer network, termed TransZero++, to refine visual features and learn accurate attribute localization for key semantic knowledge representations in ZSL. Specifically, TransZero++ employs an attribute → visual Transformer sub-net (AVT) and a visual → attribute Transformer sub-net (VAT) to learn attribute-based visual features and visual-based attribute features, respectively. By further introducing feature-level and prediction-level semantical collaborative losses, the two attribute-guided transformers teach each other to learn semantic-augmented visual embeddings for key semantic knowledge representations via semantical collaborative learning. Finally, the semantic-augmented visual embeddings learned by AVT and VAT are fused to conduct desirable visual-semantic interaction cooperated with class semantic vectors for ZSL classification. Extensive experiments show that TransZero++ achieves the new state-of-the-art results on three golden ZSL benchmarks and on the large-scale ImageNet dataset. The project website is available at: https://shiming-chen.github.io/TransZero-pp/TransZero-pp.html.

A Componentwise Approach to Weakly Supervised Semantic Segmentation Using Dual-Feedback Network

Recent weakly supervised semantic segmentation methods generate pseudolabels to recover the lost position information in weak labels for training the segmentation network. Unfortunately, those pseudolabels often contain mislabeled regions and inaccurate boundaries due to the incomplete recovery of position information. It turns out that the result of semantic segmentation becomes determinate to a certain degree. In this article, we decompose the position information into two components: high-level semantic information and low-level physical information, and develop a componentwise approach to recover each component independently. Specifically, we propose a simple yet effective pseudolabels updating mechanism to iteratively correct mislabeled regions inside objects to precisely refine high-level semantic information. To reconstruct low-level physical information, we utilize a customized superpixel-based random walk mechanism to trim the boundaries. Finally, we design a novel network architecture, namely, a dual-feedback network (DFN), to integrate the two mechanisms into a unified model. Experiments on benchmark datasets show that DFN outperforms the existing state-of-the-art methods in terms of intersection-over-union (mIoU).

Sparse Additive Machine With the Correntropy-Induced Loss

Sparse additive machines (SAMs) have shown competitive performance on variable selection and classification in high-dimensional data due to their representation flexibility and interpretability. However, the existing methods often employ the unbounded or nonsmooth functions as the surrogates of 0-1 classification loss, which may encounter the degraded performance for data with outliers. To alleviate this problem, we propose a robust classification method, named SAM with the correntropy-induced loss (CSAM), by integrating the correntropy-induced loss (C-loss), the data-dependent hypothesis space, and the weighted lq,1 -norm regularizer ( q ≥ 1 ) into additive machines. In theory, the generalization error bound is estimated via a novel error decomposition and the concentration estimation techniques, which shows that the convergence rate O(n-1/4) can be achieved under proper parameter conditions. In addition, the theoretical guarantee on variable selection consistency is analyzed. Experimental evaluations on both synthetic and real-world datasets consistently validate the effectiveness and robustness of the proposed approach.

A215 EVALUATING THE COMPARABILITY OF CARE FOR PERSONS ADMITTED TO TORONTO AREA HOSPITALS WITH ACUTE SEVERE ULCERATIVE COLITIS

Background

Approximately 20% of patients with ulcerative colitis will experience an acute severe exacerbation requiring hospitalization. Acute severe ulcerative colitis (ASUC) is a medical emergency associated with significant morbidity and a mortality rate of 1%. Timely initiation of treatment and assessment of clinical response is critical in the management of ASUC. With an aim to reduce treatment variability and improve outcomes, multiple gastrointestinal societies have published guidelines highlighting recommendations for optimal care in ASUC. It remains unclear how closely these guidelines are implemented in clinical practice. Measuring adherence to these recommended processes of care may act as a surrogate measure for quality of care and a way to indirectly evaluate outcomes in the management of patients with ASUC. Studies have shown that even amongst experienced providers practice pattern variability exists. Identifying significant variations in the management of patients with ASUC will highlight where improvement in guideline dissemination and greater adherence is required.

Purpose

We sought to evaluate how quality of care indicators varied across 7 hospital sites for patients admitted ASUC in the Greater Toronto Area.

Method

Using GEMINI, a research collaborative that collects and analyses data from inpatient admissions at 7 Toronto area hospitals, we identified patients admitted to hospital with ASUC from June 2016-December 2019. Hospital sites were further categorized into 3 hospital types; 1 IBD specialty centre (ISC), 3 other academic centres (AC) and 3 community centres (CC). Process measures assessed included proportion tested for C-reactive protein at baseline and following treatment initiation, duration of corticosteroid use, timing and initiation of biologic agents, rates of venous thromboembolism prophylaxis and opioid use. Outcome measures included hospital length of stay, rates of colectomy and mortality.

Result(s)

765 hospitalizations were included in the study; 320 occurring at ISC, 308 at AC and 137 at CC. Corticosteroid use on admission were highest at the ISC at 78% compared to 64% at AC and 63% at CC (p <0.001). Among those who received steroids on admission, 47% of patients remained on intravenous corticosteroids for at least 5 days in the ISC compared to 39% in AC and 75% in CC (p< 0.001). Initiation of biologic rescue therapy was highest at the ISC occurring in 37% of hospitalizations compared to 22% in AC and 23% in CC (p<0.001). In addition, VTE prophylaxis rates were highest at the ISC at 83% followed by 60% in AC and 45% in CC (p<0.001). Rates of colectomy were highest at ISC (12% of hospitalizations vs. 7% in AC).

Conclusion(s)

Greater adherence to indicators of quality of care were seen at the ISC compared to ACs and CCs, although patient outcomes assessed were not clearly different between sites. Further strategies are required to improve adherence to markers of quality care for patients admitted with ASUC.

Please acknowledge all funding agencies by checking the applicable boxes below

None

Disclosure of Interest

None Declared

Region-related focal loss for 3D brain tumor MRI segmentation

BACKGROUND

In the brain tumor magnetic resonance image (MRI) segmentation, although the 3D convolution networks (CNNs) has achieved state-of-the-art results, the class and hard-voxel imbalances in the 3D images have not been well addressed. Voxel independent losses are dependent on the setting of class weights for the class imbalance issue, and are hard to assign each class equally. Region-related losses cannot correctly focus on hard voxels dynamically and not be robust to misclassification of small structures. Meanwhile, repeatedly training on the additional hard samples augmented by existing methods would bring more class imbalance, overfitting and incorrect knowledge learning to the model.

PURPOSE

A novel region-related loss with balanced dynamic weighting while alleviating the sensitivity to small structures is necessary. In addition, we need to increase the diversity of hard samples in the training to improve the performance of model.

METHODS

The proposed Region-related Focal Loss (RFL) reshapes standard Dice Loss (DL) by up-weighting the loss assigned to hard-classified voxels. Compared to DL, RFL adaptively modulate its gradient with an invariant focalized point that voxels with lower-confidence than it would achieve a larger gradient, and higher-confidence voxels would get a smaller gradient. Meanwhile, RFL can adjust the parameters to set where and how much the network is focused. In addition, an Intra-classly Transformed Augmentation network (ITA-NET) is proposed to increase the diversity of hard samples, in which the 3D registration network and intra-class transfer layer are used to transform the shape and intensity respectively. A selective hard sample mining(SHSM) strategy is used to train the ITA-NET for avoiding excessive class imbalance. Source code (in Tensorflow) is available at: https://github.com/lb-whu/RFL_ITA.

RESULTS

The experiments are carried out on public data set: Brain Tumor Segmentation Challenge 2020 (BratS2020). Experiments with BraTS2020 online validation set show that proposed methods achieve an average Dice scores of 0.905, 0.821, and 0.806 for whole tumor (WT), tumor core (TC) and enhancing tumor (ET), respectively. Compared with DL (baseline), the proposed RFL significantly improves the Dice scores by an average of 1%, and for the small region ET it can even increase by 3%. And the proposed method combined with ITA-NET improves the Dice scores of ET and TC by 5% and 3% respectively.

CONCLUSIONS

The proposed RFL can converge with a invariant focalized point in the training of segmentation network, thus effectively alleviating the hard-voxel imbalance in brain tumor MRI segmentation. The negative region term of RFL can effectively reduce the sensitivity of the segmentation model to the misclassification of small structures. The proposed ITA-NET can increase the diversity of hard samples by transforming their shape and transfer their intra-class intensity, thereby effectively improving the robustness of the segmentation network to hard samples.

Learning Performance of Weighted Distributed Learning With Support Vector Machines

The divide-and-conquer strategy is a very effective method of dealing with big data. Noisy samples in big data usually have a great impact on algorithmic performance. In this article, we introduce Markov sampling and different weights for distributed learning with the classical support vector machine (cSVM). We first estimate the generalization error of weighted distributed cSVM algorithm with uniformly ergodic Markov chain (u.e.M.c.) samples and obtain its optimal convergence rate. As applications, we obtain the generalization bounds of weighted distributed cSVM with strong mixing observations and independent and identically distributed (i.i.d.) samples, respectively. We also propose a novel weighted distributed cSVM based on Markov sampling (DM-cSVM). The numerical studies of benchmark datasets show that the DM-cSVM algorithm not only has better performance but also has less total time of sampling and training compared to other distributed algorithms.

Cloning, genome structure and expression analysis of MHC class I gene in Korean quail

1. The major histocompatibility complex (MHC) is a highly polymorphic region of the genome essential to immune responses and animal health. However, avian MHC genetic structure is different from that of mammals. In this study, the structure and expression of Korean quail MHC class I gene was analysed.2. The quail MHC gene consisted of eight exons and seven introns. The open reading frame of the cDNA was 353 amino acids, and the molecular weight was about 38.91 kDa. Exons 1 and 2 coded for leading peptides and alpha 1 regions, respectively. Exons 3 and 4 encoded alpha 2 and alpha 3 regions. Exons 5 to 8 coded for connecting peptides and transmembrane regions/cytoplasmic regions (TM/CY). The Korean quail MHC class I amino acid sequence shared 87% to 99% homology with Japanese quail and 71% to 75% with chicken. The amino acid shared 40% and 43% homology with humans and mice, respectively.3. Real-time quantitative PCR showed that MHC-I was highly expressed in immune tissues such as the bursa of Fabricius. Moreover, the constructed evolutionary tree was consistent with accepted evolutionary pathways.4. MHC-I is closely related to the host's immune system, and these findings may help to better understand the role of Korean quail MHC-I in the immune system.

IAS-NET: Joint intraclassly adaptive GAN and segmentation network for unsupervised cross-domain in neonatal brain MRI segmentation

PURPOSE

In neonatal brain magnetic resonance image (MRI) segmentation, the model we trained on the training set (source domain) often performs poorly in clinical practice (target domain). As the label of target-domain images is unavailable, this cross-domain segmentation needs unsupervised domain adaptation (UDA) to make the model adapt to the target domain. However, the shape and intensity distribution of neonatal brain MRI images across the domains are largely different from adults'. Current UDA methods aim to make synthesized images similar to the target domain as a whole. But it is impossible to synthesize images with intraclass similarity because of the regional misalignment caused by the cross-domain difference. This will result in generating intraclassly incorrect intensity information from target-domain images. To address this issue, we propose an IAS-NET (joint intraclassly adaptive generative adversarial network (GAN) (IA-NET) and segmentation) framework to bridge the gap between the two domains for intraclass alignment.

METHODS

Our proposed IAS-NET is an elegant learning framework that transfers the appearance of images across the domains from both image and feature perspectives. It consists of the proposed IA-NET and a segmentation network (S-NET). The proposed IA-NET is a GAN-based adaptive network that contains one generator (including two encoders and one shared decoder) and four discriminators for cross-domain transfer. The two encoders are implemented to extract original image, mean, and variance features from source and target domains. The proposed local adaptive instance normalization algorithm is used to perform intraclass feature alignment to the target domain in the feature-map level. S-NET is a U-net structure network that is used to provide semantic constraint by a segmentation loss for the training of IA-NET. Meanwhile, it offers pseudo-label images for calculating intraclass features of the target domain. Source code (in Tensorflow) is available at https://github.com/lb-whu/RAS-NET/.

RESULTS

Extensive experiments are carried out on two different data sets (NeoBrainS12 and dHCP), respectively. There exist great differences in the shape, size, and intensity distribution of magnetic resonance (MR) images in the two databases. Compared to baseline, we improve the average dice score of all tissues on NeoBrains12 by 6% through adaptive training with unlabeled dHCP images. Besides, we also conduct experiments on dHCP and improved the average dice score by 4%. The quantitative analysis of the mean and variance of the synthesized images shows that the synthesized image by the proposed is closer to the target domain both in the full brain or within each class than that of the compared methods.

CONCLUSIONS

In this paper, the proposed IAS-NET can improve the performance of the S-NET effectively by its intraclass feature alignment in the target domain. Compared to the current UDA methods, the synthesized images by IAS-NET are more intraclassly similar to the target domain for neonatal brain MR images. Therefore, it achieves state-of-the-art results in the compared UDA models for the segmentation task.

Modal-Regression-Based Structured Low-Rank Matrix Recovery for Multiview Learning

Low-rank Multiview Subspace Learning (LMvSL) has shown great potential in cross-view classification in recent years. Despite their empirical success, existing LMvSL-based methods are incapable of handling well view discrepancy and discriminancy simultaneously, which, thus, leads to performance degradation when there is a large discrepancy among multiview data. To circumvent this drawback, motivated by the block-diagonal representation learning, we propose structured low-rank matrix recovery (SLMR), a unique method of effectively removing view discrepancy and improving discriminancy through the recovery of the structured low-rank matrix. Furthermore, recent low-rank modeling provides a satisfactory solution to address the data contaminated by the predefined assumptions of noise distribution, such as Gaussian or Laplacian distribution. However, these models are not practical, since complicated noise in practice may violate those assumptions and the distribution is generally unknown in advance. To alleviate such a limitation, modal regression is elegantly incorporated into the framework of SLMR (termed MR-SLMR). Different from previous LMvSL-based methods, our MR-SLMR can handle any zero-mode noise variable that contains a wide range of noise, such as Gaussian noise, random noise, and outliers. The alternating direction method of multipliers (ADMM) framework and half-quadratic theory are used to optimize efficiently MR-SLMR. Experimental results on four public databases demonstrate the superiority of MR-SLMR and its robustness to complicated noise.

Multiset Feature Learning for Highly Imbalanced Data Classification

With the expansion of data, increasing imbalanced data has emerged. When the imbalance ratio (IR) of data is high, most existing imbalanced learning methods decline seriously in classification performance. In this paper, we systematically investigate the highly imbalanced data classification problem, and propose an uncorrelated cost-sensitive multiset learning (UCML) approach for it. Specifically, UCML first constructs multiple balanced subsets through random partition, and then employs the multiset feature learning (MFL) to learn discriminant features from the constructed multiset. To enhance the usability of each subset and deal with the non-linearity issue existed in each subset, we further propose a deep metric based UCML (DM-UCML) approach. DM-UCML introduces the generative adversarial network technique into the multiset constructing process, such that each subset can own similar distribution with the original dataset. To cope with the non-linearity issue, DM-UCML integrates deep metric learning with MFL, such that more favorable performance can be achieved. In addition, DM-UCML designs a new discriminant term to enhance the discriminability of learned metrics. Experiments on eight traditional highly class-imbalanced datasets and two large-scale datasets indicate that: the proposed approaches outperform state-of-the-art highly imbalanced learning methods and are more robust to high IR.

Knockdown of LINC00461 inhibits cell proliferation and induces apoptosis in gastric cancer by targeting LSD1

OBJECTIVE

To uncover the function of LINC00461 in regulating cellular behaviors of gastric cancer (GC) via targeting LSD1.

PATIENTS AND METHODS

LINC00461 level in GC tissues with different tumor node metastasis (TNM) staging and lymphatic metastasis statues was determined by quantitative Real Time-Polymerase Chain Reaction (qRT-PCR). In vitro influences of LINC00461 on proliferative and apoptotic rates were evaluated in AGS and SGC-7901 cells. The interaction between LINC00461 and LSD1 was explored by RNA immunoprecipitation (RIP) assay and qRT-PCR. Finally, the potential role of LSD1 in the proliferative ability of GC cells mediated by LINC00461 was assessed.

RESULTS

LINC00461 level was higher in GC tissues relative to matched control ones. It was positively correlated to TNM staging and lymphatic metastasis of GC. Knockdown of LINC00461 markedly attenuated viability and the proliferative ability of AGS and SGC-7901 cells, but induced apoptosis. RIP assay demonstrated the interaction between LINC00461 and LSD1. Moreover, LSD1 could reverse the regulatory effect of LINC00461 on the proliferative ability of GC cells.

CONCLUSIONS

LINC00461 is upregulated in GC, which is positively related to TNM staging and lymphatic metastasis. LINC00461 mediates proliferation and apoptosis of GC cells, thereafter aggravating the progression of GC.

LncRNA PAPAS aggravates the progression of gastric cancer through regulating miRNA-188-5p

OBJECTIVE

To uncover the biological effect of long non-coding RNA (lncRNA) PAPAS on the progression of gastric cancer (GC) by mediating microRNA-188-5p (miRNA-188-5p) level.

PATIENTS AND METHODS

The relative level of PAPAS was determined in GC tissues and cell lines by quantitative Real Time-Polymerase Chain Reaction (qRT-PCR). The Kaplan-Meier method was introduced to assess the prognostic potential of PAPAS in the overall survival of GC patients. Regulatory effects of PAPAS on proliferative, migratory, and invasive abilities of HGC-27 and AGS cells were detected by cell counting kit-8 (CCK-8), transwell, and wound closure assay, respectively. Subsequently, the binding relation between PAPAS and miRNA-188-5p was verified by the Dual-luciferase reporter gene assay. Correlation between expression levels of PAPAS and miRNA-188-5p in GC tissues was explored. Finally, rescue experiments were conducted to uncover the role of PAPAS/miRNA-188-5p axis in the progression of GC.

RESULTS

PAPAS was upregulated in GC tissues and cell lines compared to controls. GC patients expressing a high level of PAPAS suffered worse prognosis relative to those with low level. The silence of PAPAS remarkably attenuated proliferative, migratory, and invasive abilities of HGC-27 cells. Overexpression of PAPAS in AGS cells obtained the opposite trends. MiRNA-188-5p was the direct target of PAPAS, which was negatively regulated by PAPAS. MiRNA-188-5p was able to reverse the regulatory effects of PA-PAS on proliferative, migratory, and invasive abilities of GC cells.

CONCLUSIONS

LncRNA PAPAS is upregulated in GC and closely related to lymphatic metastasis, distant metastasis, and poor prognosis of GC patients. PAPAS aggravate the malignant progression of GC by negatively regulating the miRNA-188-5p level.

Mycoplasma pneumoniae lipids license TLR-4 for activation of NLRP3 inflammasome and autophagy to evoke a proinflammatory response

Mycoplasma pneumoniae is an obligate pathogen that causes pneumonia, tracheobronchitis, pharyngitis and asthma in humans. It is well recognized that membrane lipoproteins are immunostimulants exerting as lipopolysaccharides (LPS) and play a crucial role in the pathogenesis of inflammatory responses upon M. pneumoniae infection. Here, we report that the M. pneumoniae-derived lipids are another proinflammatory agents. Using an antibody-neutralizing assay, RNA interference or specific inhibitors, we found that Toll-like receptor 4 (TLR-4) is essential for M. pneumoniae lipid-induced tumour necrosis factor (TNF)-α and interleukin (IL)-1β production. We also demonstrate that NLR family pyrin domain containing 3 inflammasome (NLRP3) inflammasome, autophagy and nuclear factor kappa B (NF-κB)-dependent pathways are critical for the secretion of proinflammatory cytokines, while inhibition of TLR-4 significantly abrogates these events. Further characterization revealed that autophagy-mediated inflammatory responses involved the activation of NF-κB. In addition, the activation of NF-κB promoted lipid-induced autophagosome formation, as revealed by assays using pharmacological inhibitors, 3-methyladenine (3-MA) and Bay 11-7082, or silencing of atg5 and beclin-1. These findings suggest that, unlike the response to lipoprotein stimulation, the inflammation in response to M. pneumoniae lipids is mediated by the TLR-4 pathway, which subsequently initiates the activation of NLRP3 inflammasome and formation of a positive feedback loop between autophagy and NF-κB signalling cascade, ultimately promoting TNF-α and Il-1β production in macrophages.

Multi-instance Deep Learning of Ultrasound Imaging Data for Pattern Classification of Congenital Abnormalities of the Kidney and Urinary Tract in Children

OBJECTIVE

To reliably and quickly diagnose children with posterior urethral valves (PUV), we developed a multi-instance deep learning method to automate image analysis.

METHODS

We built a robust pattern classifier to distinguish 86 children with PUV from 71 children with mild unilateral hydronephrosis based on ultrasound images (3504 in sagittal view and 2558 in transverse view) obtained during routine clinical care.

RESULTS

The multi-instance deep learning classifier performed better than classifiers built on either single sagittal images or single transverse images. Particularly, the deep learning classifiers built on single images in the sagittal view and single images in the transverse view obtained area under the receiver operating characteristic curve (AUC) values of 0.796 ± 0.064 and 0.815 ± 0.071, respectively. AUC values of the multi-instance deep learning classifiers built on images in the sagittal and transverse views with mean pooling operation were 0.949 ± 0.035 and 0.954 ± 0.033, respectively. The multi-instance deep learning classifiers built on images in both the sagittal and transverse views with a mean pooling operation obtained an AUC of 0.961 ± 0.026 with a classification rate of 0.925 ± 0.060, specificity of 0.986 ± 0.032, and sensitivity of 0.873 ± 0.120, respectively. Discriminative regions of the kidney located using classification activation mapping demonstrated that the deep learning techniques could identify meaningful anatomical features from ultrasound images.

CONCLUSION

The multi-instance deep learning method provides an automatic and accurate means to extract informative features from ultrasound images and discriminate infants with PUV from male children with unilateral hydronephrosis.

Multiview Hybrid Embedding: A Divide-and-Conquer Approach

We present a novel cross-view classification algorithm where the gallery and probe data come from different views. A popular approach to tackle this problem is the multiview subspace learning (MvSL) that aims to learn a latent subspace shared by multiview data. Despite promising results obtained on some applications, the performance of existing methods deteriorates dramatically when the multiview data is sampled from nonlinear manifolds or suffers from heavy outliers. To circumvent this drawback, motivated by the Divide-and-Conquer strategy, we propose multiview hybrid embedding (MvHE), a unique method of dividing the problem of cross-view classification into three subproblems and building one model for each subproblem. Specifically, the first model is designed to remove view discrepancy, whereas the second and third models attempt to discover the intrinsic nonlinear structure and to increase the discriminability in intraview and interview samples, respectively. The kernel extension is conducted to further boost the representation power of MvHE. Extensive experiments are conducted on four benchmark datasets. Our methods demonstrate the overwhelming advantages against the state-of-the-art MvSL-based cross-view classification approaches in terms of classification accuracy and robustness.

Computer-Aided Diagnosis of Congenital Abnormalities of the Kidney and Urinary Tract in Children Using a Multi-Instance Deep Learning Method Based on Ultrasound Imaging Data

Ultrasound images are widely used for diagnosis of congenital abnormalities of the kidney and urinary tract (CAKUT). Since a typical clinical ultrasound image captures 2D information of a specific view plan of the kidney and images of the same kidney on different planes have varied appearances, it is challenging to develop a computer aided diagnosis tool robust to ultrasound images in different views. To overcome this problem, we develop a multi-instance deep learning method for distinguishing children with CAKUT from controls based on their clinical ultrasound images, aiming to automatic diagnose the CAKUT in children based on ultrasound imaging data. Particularly, a multi-instance deep learning method was developed to build a robust pattern classifier to distinguish children with CAKUT from controls based on their ultrasound images in sagittal and transverse views obtained during routine clinical care. The classifier was built on imaging features derived using transfer learning from a pre-trained deep learning model with a mean pooling operator for fusing instance-level classification results. Experimental results have demonstrated that the multi-instance deep learning classifier performed better than classifiers built on either individual sagittal slices or individual transverse slices.

Automatic kidney segmentation in ultrasound images using subsequent boundary distance regression and pixelwise classification networks

It remains challenging to automatically segment kidneys in clinical ultrasound (US) images due to the kidneys' varied shapes and image intensity distributions, although semi-automatic methods have achieved promising performance. In this study, we propose subsequent boundary distance regression and pixel classification networks to segment the kidneys automatically. Particularly, we first use deep neural networks pre-trained for classification of natural images to extract high-level image features from US images. These features are used as input to learn kidney boundary distance maps using a boundary distance regression network and the predicted boundary distance maps are classified as kidney pixels or non-kidney pixels using a pixelwise classification network in an end-to-end learning fashion. We also adopted a data-augmentation method based on kidney shape registration to generate enriched training data from a small number of US images with manually segmented kidney labels. Experimental results have demonstrated that our method could automatically segment the kidney with promising performance, significantly better than deep learning-based pixel classification networks.

Proteomic and Transcriptomic Changes in Hibernating Grizzly Bears Reveal Metabolic and Signaling Pathways that Protect against Muscle Atrophy

Muscle atrophy is a physiological response to disuse and malnutrition, but hibernating bears are largely resistant to this phenomenon. Unlike other mammals, they efficiently reabsorb amino acids from urine, periodically activate muscle contraction, and their adipocytes differentially responds to insulin. The contribution of myocytes to the reduced atrophy remains largely unknown. Here we show how metabolism and atrophy signaling are regulated in skeletal muscle of hibernating grizzly bear. Metabolic modeling of proteomic changes suggests an autonomous increase of non-essential amino acids (NEAA) in muscle and treatment of differentiated myoblasts with NEAA is sufficient to induce hypertrophy. Our comparison of gene expression in hibernation versus muscle atrophy identified several genes differentially regulated during hibernation, including Pdk4 and Serpinf1. Their trophic effects extend to myoblasts from non-hibernating species (including C. elegans), as documented by a knockdown approach. Together, these changes reflect evolutionary favored adaptations that, once translated to the clinics, could help improve atrophy treatment.

Multi-instance Deep Learning with Graph Convolutional Neural Networks for Diagnosis of Kidney Diseases Using Ultrasound Imaging

Ultrasound imaging (US) is commonly used in nephrology for diagnostic studies of the kidneys and lower urinary tract. However, it remains challenging to automate the disease diagnosis based on clinical 2D US images since they provide partial anatomic information of the kidney and the 2D images of the same kidney may have heterogeneous appearance. To overcome this challenge, we develop a novel multi-instance deep learning method to build a robust classifier by treating multiple 2D US images of each individual subject as multiple instances of one bag. Particularly, we adopt convolutional neural networks (CNNs) to learn instance-level features from 2D US kidney images and graph convolutional networks (GCNs) to further optimize the instance-level features by exploring potential correlation among instances of the same bag. We also adopt a gated attention-based MIL pooling to learn bag-level features using full-connected neural networks (FCNs). Finally, we integrate both instance-level and bag-level supervision to further improve the bag-level classification accuracy. Ablation studies and comparison results have demonstrated that our method could accurately diagnose kidney diseases using ultrasound imaging, with better performance than alternative state-of-the-art multi-instance deep learning methods.

Stroke in pregnancy

Stroke is uncommon among young adults. However, the incidence of stroke among young women increases with pregnancy during peripartum and postpartum periods. The relative risk of suffering from haemorrhagic stroke was three times higher than ischemic stroke during these periods when compared with antenatal period. Neuroimaging should be prioritized in order to establish diagnosis and to facilitate treatment in a patient with suspected acute stroke. Prophylaxic anticoagulants should be used in high risk patients. Treatments of acute stroke in pregnant women include anti-platelet and thrombolytic agents. Further studies should be carried as there is lack of high level of evidences to formulate clear guideline for the management of stroke during pregnancy.

FULLY-AUTOMATIC SEGMENTATION OF KIDNEYS IN CLINICAL ULTRASOUND IMAGES USING A BOUNDARY DISTANCE REGRESSION NETWORK

It remains challenging to automatically segment kidneys in clinical ultrasound images due to the kidneys' varied shapes and image intensity distributions, although semi-automatic methods have achieved promising performance. In this study, we developed a novel boundary distance regression deep neural network to segment the kidneys, informed by the fact that the kidney boundaries are relatively consistent across images in terms of their appearance. Particularly, we first use deep neural networks pre-trained for classification of natural images to extract high-level image features from ultrasound images, then these feature maps are used as input to learn kidney boundary distance maps using a boundary distance regression network, and finally the predicted boundary distance maps are classified as kidney pixels or non-kidney pixels using a pixel classification network in an end-to-end learning fashion. Experimental results have demonstrated that our method could effectively improve the performance of automatic kidney segmentation, significantly better than deep learning based pixel classification networks.

Video-based Person Re-identification by Simultaneously Learning Intra-video and Inter-video Distance Metrics

Video-based person re-identification (re-id) is an important application in practice. Since large variations exist between different pedestrian videos, as well as within each video, it's challenging to conduct re-identification between pedestrian videos. In this paper, we propose a simultaneous intra-video and inter-video distance learning (SI2DL) approach for video-based person re-id. Specifically, SI2DL simultaneously learns an intravideo distance metric and an inter-video distance metric from the training videos. The intra-video distance metric is used to make each video more compact, and the inter-video one is used to ensure that the distance between truly matching videos is smaller than that between wrong matching videos. Considering that the goal of distance learning is to make truly matching video pairs from different persons be well separated with each other, we also propose a pair separation based SI2DL (P-SI2DL). P-SI2DL aims to learn a pair of distance metrics, under which any two truly matching video pairs can be well separated. Experiments on four public pedestrian image sequence datasets show that our approaches achieve the state-of-the-art performance.

A joint space-angle regularization approach for single 4D diffusion image super-resolution

PURPOSE

Low signal-to-noise-ratio and limited scan time of diffusion magnetic resonance imaging (dMRI) in current clinical settings impede obtaining images with high spatial and angular resolution (HSAR) for a reliable fiber reconstruction with fine anatomical details. To overcome this problem, we propose a joint space-angle regularization approach to reconstruct HSAR diffusion signals from a single 4D low resolution (LR) dMRI, which is down-sampled in both 3D-space and q-space.

METHODS

Different from the existing works which combine multiple 4D LR diffusion images acquired using specific acquisition protocols, the proposed method reconstructs HSAR dMRI from only a single 4D dMRI by exploring and integrating two key priors, that is, the nonlocal self-similarity in the spatial domain as a prior to increase spatial resolution and ridgelet approximations in the diffusion domain as another prior to increase the angular resolution of dMRI. To more effectively capture nonlocal self-similarity in the spatial domain, a novel 3D block-based nonlocal means filter is imposed as the 3D image space regularization term which is accurate in measuring the similarity and fast for 3D reconstruction. To reduce computational complexity, we use the L₂ -norm instead of sparsity constraint on the representation coefficients.

RESULTS

Experimental results demonstrate that the proposed method can obtain the HSAR dMRI efficiently with approximately 2% per-voxel root-mean-square error between the actual and reconstructed HSAR dMRI.

CONCLUSION

The proposed approach can effectively increase the spatial and angular resolution of the dMRI which is independent of the acquisition protocol, thus overcomes the inherent resolution limitation of imaging systems.

$k$ -Times Markov Sampling for SVMC

Support vector machine (SVM) is one of the most widely used learning algorithms for classification problems. Although SVM has good performance in practical applications, it has high algorithmic complexity as the size of training samples is large. In this paper, we introduce SVM classification (SVMC) algorithm based on -times Markov sampling and present the numerical studies on the learning performance of SVMC with -times Markov sampling for benchmark data sets. The experimental results show that the SVMC algorithm with -times Markov sampling not only have smaller misclassification rates, less time of sampling and training, but also the obtained classifier is more sparse compared with the classical SVMC and the previously known SVMC algorithm based on Markov sampling. We also give some discussions on the performance of SVMC with -times Markov sampling for the case of unbalanced training samples and large-scale training samples.

Dynamically Modulated Mask Sparse Tracking

Visual tracking is a critical task in many computer vision applications such as surveillance and robotics. However, although the robustness to local corruptions has been improved, prevailing trackers are still sensitive to large scale corruptions, such as occlusions and illumination variations. In this paper, we propose a novel robust object tracking technique depends on subspace learning-based appearance model. Our contributions are twofold. First, mask templates produced by frame difference are introduced into our template dictionary. Since the mask templates contain abundant structure information of corruptions, the model could encode information about the corruptions on the object more efficiently. Meanwhile, the robustness of the tracker is further enhanced by adopting system dynamic, which considers the moving tendency of the object. Second, we provide the theoretic guarantee that by adapting the modulated template dictionary system, our new sparse model can be solved by the accelerated proximal gradient algorithm as efficient as in traditional sparse tracking methods. Extensive experimental evaluations demonstrate that our method significantly outperforms 21 other cutting-edge algorithms in both speed and tracking accuracy, especially when there are challenges such as pose variation, occlusion, and illumination changes.

First Direct-Detection Constraints on eV-Scale Hidden-Photon Dark Matter with DAMIC at SNOLAB

We present direct detection constraints on the absorption of hidden-photon dark matter with particle masses in the range 1.2-30  eV c^{-2} with the DAMIC experiment at SNOLAB. Under the assumption that the local dark matter is entirely constituted of hidden photons, the sensitivity to the kinetic mixing parameter κ is competitive with constraints from solar emission, reaching a minimum value of 2.2×10^{-14} at 17  eV c^{-2}. These results are the most stringent direct detection constraints on hidden-photon dark matter in the galactic halo with masses 3-12  eV c^{-2} and the first demonstration of direct experimental sensitivity to ionization signals <12  eV from dark matter interactions.

The α-β circular scanning with large range and low noise

A circular-route scanning method called α-β circular scanning is proposed and realized using sinusoidal signals with a constant phase difference of π/2. Experiments show that the circular scanning range of α-β circular scanning is 57% greater than the rectangular scanning range of raster scanning within an effective optical field of view. Moreover, the scanning speed is improved by 7.8% over raster scanning because the whole sine signal is utilized in α-β circular scanning whereas the flyback area of the saw-tooth signal needs to be discarded in raster scanning. The maximum scanning acceleration decreases by a factor of 44, drastically decreasing the high noise, which should considerably elongate the lifetime of the galvanometers while inhibiting internal vibration. The proposed α-β circular scanning technique could be used in scanning imaging, optical tweezers and laser-beam fabrication.

Super-Resolution Person Re-Identification With Semi-Coupled Low-Rank Discriminant Dictionary Learning

Person re-identification has been widely studied due to its importance in surveillance and forensics applications. In practice, gallery images are high resolution (HR), while probe images are usually low resolution (LR) in the identification scenarios with large variation of illumination, weather, or quality of cameras. Person re-identification in this kind of scenarios, which we call super-resolution (SR) person re-identification, has not been well studied. In this paper, we propose a semi-coupled low-rank discriminant dictionary learning (SLD²L) approach for SR person re-identification task. With the HR and LR dictionary pair and mapping matrices learned from the features of HR and LR training images, SLD²L can convert the features of the LR probe images into HR features. To ensure that the converted features have favorable discriminative capability and the learned dictionaries can well characterize intrinsic feature spaces of the HR and LR images, we design a discriminant term and a low-rank regularization term for SLD²L. Moreover, considering that low resolution results in different degrees of loss for different types of visual appearance features, we propose a multi-view SLD²L (MVSLD²L) approach, which can learn the type-specific dictionary pair and mappings for each type of feature. Experimental results on multiple publicly available data sets demonstrate the effectiveness of our proposed approaches for the SR person re-identification task.

Robust Object Tracking via Key Patch Sparse Representation

Many conventional computer vision object tracking methods are sensitive to partial occlusion and background clutter. This is because the partial occlusion or little background information may exist in the bounding box, which tends to cause the drift. To this end, in this paper, we propose a robust tracker based on key patch sparse representation (KPSR) to reduce the disturbance of partial occlusion or unavoidable background information. Specifically, KPSR first uses patch sparse representations to get the patch score of each patch. Second, KPSR proposes a selection criterion of key patch to judge the patches within the bounding box and select the key patch according to its location and occlusion case. Third, KPSR designs the corresponding contribution factor for the sampled patches to emphasize the contribution of the selected key patches. Comparing the KPSR with eight other contemporary tracking methods on 13 benchmark video data sets, the experimental results show that the KPSR tracker outperforms classical or state-of-the-art tracking methods in the presence of partial occlusion, background clutter, and illumination change.

Diagnostic value of recombinant Tp0821 protein in serodiagnosis for syphilis

Syphilis is a multistage sexually transmitted disease that remains a serious public health concern worldwide. The coexistence of Treponema pallidum with other closely related members of spirochaeta, such as Leptospira spp. and Borrelia burgdorferi, has complicated the serodiagnosis due to cross-reactive antigens. In this study, recombinant Tp0821 protein was expressed in Escherichia coli and purified by metal affinity chromatography. Then enzyme-linked immunosorbent assays (ELISAs) based on Tp0821 for the detection of specific antibodies were established. The relative positive rates of the IgM ELISA and the IgG ELISA were found to be 91·0 and 98·3%, respectively, when screening 578 syphilis specimens. The specificities were 94·3 and 100%, respectively, when cross-checking with serum samples obtained from 30 patients with Lyme disease, five patients with leptospirosis, and 52 uninfected controls. In addition, relative positive rates and specificities of Tp0821 for human sera were all 100% in Western blotting. When compared to the syphilis diagnostic tests commonly used in clinical settings, we found that the results of Tp0821-based ELISAs correlated well with the results of the treponemal tests, specifically the T. pallidum particle agglutination (TP-PA) test and the chemiluminescent immunoassay (CIA). Thus, these findings identify Tp0821 as a novel serodiagnostic candidate for syphilis.

SIGNIFICANCE AND IMPACT OF THE STUDY

In this study, we expressed and purified the Treponema pallidum protein Tp0821 and developed Tp0821-based enzyme-linked immunosorbent assays (ELISAs) for the detection of specific antibodies. The serodiagnostic performance of the recombinant protein was then evaluated. When compared to the results of syphilis diagnostic tests commonly used in clinical settings, we found that the reactivities of syphilitic sera with the recombinant antigen correlated well with the results of the treponemal tests, specifically the T. pallidum particle agglutination (TP-PA) test and the chemiluminescent immunoassay (CIA). Thus, the recombinant protein shows promise as a new diagnostic antigen in the ELISAs.

Daclizumab high-yield process reduced the evolution of new gadolinium-enhancing lesions to T1 black holes in patients with relapsing-remitting multiple sclerosis

BACKGROUND AND PURPOSE

In the SELECT study, treatment with daclizumab high-yield process (DAC HYP) versus placebo reduced the frequency of gadolinium-enhancing (Gd(+) ) lesions in patients with relapsing-remitting multiple sclerosis (RRMS). The objective of this post hoc analysis of SELECT was to evaluate the effect of DAC HYP on the evolution of new Gd(+) lesions to T1 hypointense lesions (T1 black holes).

METHODS

SELECT was a randomized double-blind study of subcutaneous DAC HYP 150 or 300 mg or placebo every 4 weeks. Magnetic resonance imaging (MRI) scans were performed at baseline and weeks 24, 36 and 52 in all patients and monthly between weeks 4 and 20 in a subset of patients. MRI scans were evaluated for new Gd(+) lesions that evolved to T1 black holes at week 52. Data for the DAC HYP groups were pooled for analysis.

RESULTS

Daclizumab high-yield process reduced the number of new Gd(+) lesions present at week 24 (P = 0.005) or between weeks 4 and 20 (P = 0.014) that evolved into T1 black holes at week 52 versus placebo. DAC HYP treatment also reduced the percentage of patients with Gd(+) lesions evolving to T1 black holes versus placebo.

CONCLUSIONS

Treatment with DAC HYP reduced the evolution of Gd(+) lesions to T1 black holes versus placebo, suggesting that inflammatory lesions that evolved during DAC HYP treatment are less destructive than those evolving during placebo treatment.

Connected Component Model for Multi-Object Tracking

In multi-object tracking, it is critical to explore the data associations by exploiting the temporal information from a sequence of frames rather than the information from the adjacent two frames. Since straightforwardly obtaining data associations from multi-frames is an NP-hard multi-dimensional assignment (MDA) problem, most existing methods solve this MDA problem by either developing complicated approximate algorithms, or simplifying MDA as a 2D assignment problem based upon the information extracted only from adjacent frames. In this paper, we show that the relation between associations of two observations is the equivalence relation in the data association problem, based on the spatial-temporal constraint that the trajectories of different objects must be disjoint. Therefore, the MDA problem can be equivalently divided into independent subproblems by equivalence partitioning. In contrast to existing works for solving the MDA problem, we develop a connected component model (CCM) by exploiting the constraints of the data association and the equivalence relation on the constraints. Based upon CCM, we can efficiently obtain the global solution of the MDA problem for multi-object tracking by optimizing a sequence of independent data association subproblems. Experiments on challenging public data sets demonstrate that our algorithm outperforms the state-of-the-art approaches.

Multiobjective Optimization of Linear Cooperative Spectrum Sensing: Pareto Solutions and Refinement

In linear cooperative spectrum sensing, the weights of secondary users and detection threshold should be optimally chosen to minimize missed detection probability and to maximize secondary network throughput. Since these two objectives are not completely compatible, we study this problem from the viewpoint of multiple-objective optimization. We aim to obtain a set of evenly distributed Pareto solutions. To this end, here, we introduce the normal constraint (NC) method to transform the problem into a set of single-objective optimization (SOO) problems. Each SOO problem usually results in a Pareto solution. However, NC does not provide any solution method to these SOO problems, nor any indication on the optimal number of Pareto solutions. Furthermore, NC has no preference over all Pareto solutions, while a designer may be only interested in some of them. In this paper, we employ a stochastic global optimization algorithm to solve the SOO problems, and then propose a simple method to determine the optimal number of Pareto solutions under a computational complexity constraint. In addition, we extend NC to refine the Pareto solutions and select the ones of interest. Finally, we verify the effectiveness and efficiency of the proposed methods through computer simulations.

Isolation and characterization of a chalcone isomerase gene promoter from potato cultivars

Chalcone isomerase (CHI) is a key enzyme involved in anthocyanin metabolism. Previous research on CHI has mainly focused on cDNA cloning and gene expression. In the current study, the 1425-bp potato CHI promoter (PCP) was isolated from four potato cultivars (Heijingang, Zhongshu 7, Désirée, and Favorita) using PCR and DNA sequencing. The PCP contained many cis-regulatory elements (CREs) related to anthocyanin metabolism, tissue specificity, light response, stress, and hormone induction. Of the PCP CREs identified, 19 were common to those found in the higher plants examined, based on plant CRE databases. Multiple sequence alignment showed six single nucleotide variation sites in PCP among the potato cultivars examined, resulting in changes in the number of CREs connected with tissue specificity, anthocyanin metabolism, and light response. The 665-bp PCP fragments from Favorita and 1425-bp PCP fragments from Heijingang were used to construct plant expression vectors, which may be a useful tool for biological engineering. A transient expression assay demonstrated that the two PCP fragments from Heijingang could direct the expression of a green fluorescent protein gene in onion epidermis and a β-glucuronidase gene in all potato tuber tissues with different colors, suggesting that the single nucleotide variation in the PCP did not affect its activity, and that silencing of the CHI gene in Favorita may be attributed to other regulatory factors.

Robust Nonnegative Patch Alignment for Dimensionality Reduction

Dimensionality reduction is an important method to analyze high-dimensional data and has many applications in pattern recognition and computer vision. In this paper, we propose a robust nonnegative patch alignment for dimensionality reduction, which includes a reconstruction error term and a whole alignment term. We use correntropy-induced metric to measure the reconstruction error, in which the weight is learned adaptively for each entry. For the whole alignment, we propose locality-preserving robust nonnegative patch alignment (LP-RNA) and sparsity-preserviing robust nonnegative patch alignment (SP-RNA), which are unsupervised and supervised, respectively. In the LP-RNA, we propose a locally sparse graph to encode the local geometric structure of the manifold embedded in high-dimensional space. In particular, we select large p -nearest neighbors for each sample, then obtain the sparse representation with respect to these neighbors. The sparse representation is used to build a graph, which simultaneously enjoys locality, sparseness, and robustness. In the SP-RNA, we simultaneously use local geometric structure and discriminative information, in which the sparse reconstruction coefficient is used to characterize the local geometric structure and weighted distance is used to measure the separability of different classes. For the induced nonconvex objective function, we formulate it into a weighted nonnegative matrix factorization based on half-quadratic optimization. We propose a multiplicative update rule to solve this function and show that the objective function converges to a local optimum. Several experimental results on synthetic and real data sets demonstrate that the learned representation is more discriminative and robust than most existing dimensionality reduction methods.