The missing of graph attributes poses a significant challenge in graph representation learning. Some existing graph attribute completion methods adopt the shared-space hypothesis or employ end-to-end frameworks to perform single-attribute imputation. However, these models can only generate one single attribute with a few specific patterns that either adhere to prior knowledge or are optimal for downstream tasks, making it difficult to capture the full range of variations in the target attribute distribution. This limitation negatively impacts the model's generalizability and efficiency. Therefore, to address this issue, we proposed a new method based on a graph denoising diffusion model, called Multi-attribute Imputation Graph Denoising Diffusion Model (MIGDiff), which can generate multiple high-quality attributes. Specifically, it employs a Dual-source Auto-encoder on existing attributes and graph topology to extract reliable knowledge, which serves as a condition for training the diffusion module. Within diffusion, noise is added to the structural embeddings of nodes without attributes in the forward process. In the reverse process, a Structure-aware Denoising Network is devised to integrate feature and structural information via an attention mechanism and to perform neighbor-guided refinement based on graph connectivity, thereby enhancing denoising and accurately recovering missing attributes while effectively maintaining structural consistency and distributional fidelity. During generation, multiple initial values are sampled to produce diverse attribute imputations, avoiding focusing on a few easy-to-learn patterns. Extensive experiments conducted on four public datasets highlight the state-of-the-art performance of MIGDiff in both attribute imputation and node classification tasks.

Graph contrastive learning (GCL) aims to learn representations by bringing semantically similar graphs closer and pushing dissimilar ones farther apart without label supervision. Hard negatives, which refer to graphs that have different labels but similar embeddings to the target graph, play a key role in improving representation discrimination. However, current methods that generate both high-quality positives and hard negatives face two challenges: (1) Hard negative sample generation often suffers from class imbalance, resulting in unequal attention across classes and reduced discriminative power in the learned representations. (2) The typical binary positive sample generation approach, which divides the graph into important and unimportant semantic regions, overlooks regions that negatively impact semantics and mislead model predictions. To address these issues, we introduce a novel method named BalanceGCL, which enhance graph contrastive learning with balanced hard negatives and fine-grained semantic-aware positives. BalanceGCL comprises two modules: Balanced Hard Negative graphs generation (BHN) and Fine-grained Semantic-aware Positive graphs generation (FSP). Inspired by the counterfactual mechanism, BHN generates balanced hard negatives that remain structurally similar to the original graph while inducing a controlled semantic shift. To ensure class balance, BHN iteratively constructs one hard negative sample for each class, ensuring an even distribution of negative samples across all alternative categories. FSP leverages the semantic differences between original graphs and balanced hard negatives to identify positively contributing, negatively contributing, and unimportant regions. By enhancing the influence of positive contributors, suppressing negative ones, and perturbing unimportant areas, it generates more reliable and semantically complete positive samples. The proposed method outperforms state-of-the-art GCL techniques across 14 datasets in graph classification and transfer learning tasks, demonstrating its effectiveness in tackling class imbalance and identifying fine-grained semantic-aware regions.

Pre-trained vision-language (V-L) models exhibit promising performance across various general-domain tasks. However, they fall short in medical pathology due to the critical need for fine-grained semantic alignment, which is essential for distinguishing subtle visual patterns across categories. This limitation is not merely due to domain gaps but stems from the inability to capture detailed, pathology-specific semantics. Previous efforts leveraging large language models (LLMs) or cross-modal training often introduce redundant or ambiguous cues, ultimately weakening generalization. To explicitly enhance fine-grained alignment, we propose a Variational Distillation framework tailored for Medical Pathology V-L models. This method introduces a dual-loop optimization mechanism that jointly distills and aligns semantic signals from both textual inputs and external LLM knowledge. Specifically, we use variational latent distributions to model semantic ambiguity and apply a KL-based loss to reduce differences between signals. This encourages the model to retain robust and generalizable features, enabling improved sensitivity to subtle semantic variations critical in pathology image understanding. During cross-modal alignment, the proposed method further amplifies modality-shared semantics while suppressing modality-specific noise and task-irrelevant factors, yielding more precise and pathology-aware image-text matching. Extensive experiments on five pathology benchmarks across three settings, including class generalization, few-shot learning, and cross-organ transfer, demonstrate that the proposed method consistently outperforms the existing approaches.

The integration of biological networks provides crucial support for accurate gene function prediction, a task that aims to assign genes to corresponding functional categories through computational methods. However, existing approaches struggle with multi-source heterogeneous networks due to their limited ability to capture complex nonlinear dependencies. Contrastive learning, which captures data distributions by measuring similarities and dissimilarities between samples, can generate semantically rich feature representations, offering a new approach to address the aforementioned issues. In this work, we propose EPILOGUE, a multi-view graph contrastive learning framework for gene function prediction. By integrating graph neural networks with contrastive learning, EPILOGUE enables the extraction of high-quality, discriminative gene representations for accurate functional annotation. Additionally, protein sequences are used as node features, offering biological information beyond network topology and supporting the learning of comprehensive semantic representations. Experiments on yeast and human datasets from the STRING database demonstrate that EPILOGUE outperforms nine state-of-the-art methods across six evaluation metrics, validating its effectiveness in learning semantically rich representations for gene function annotation.

Spatial transcriptomics technologies carry out advanced sequencing analysis of molecular profiles with a spatial context, providing multi-source information essential for elucidating biological regulatory mechanisms. Nonetheless, it poses challenges in the integration of raw spatial coordinates with high-dimensional gene expression profiles in their native feature space. While spatial-aware methods effectively aggregate molecular information from local spatial neighborhoods, they fail to explore the long-range relationships associated with gene expression data. To address this issue, this paper introduces a novel approach termed GraphSTAR that encodes both spatial and gene expression data into undirected graphs, characterizing the local spatial proximity and global transcriptional similarity, respectively. Through a graph aggregation process, GraphSTAR integrates these diverse data sources within a joint graph structure, effectively modeling both local neighborhood relationships and long-range functional associations. Subsequently, a reassembled graph neural network is established by incorporating the graph aggregation into the feed-forward propagation using proximal operators, progressively refining spatial-informed latent representation to decipher spatial expression patterns of genes. Extensive experiments on benchmark datasets demonstrate that GraphSTAR outperforms state-of-the-art methods in both spatial domain identification and cell-type annotation tasks.

Medical AI models have achieved remarkable progress in various tasks, including medical image classification and segmentation. However, a critical issue frequently overlooked in clinical applications is the fairness of these models across different subgroups. Existing data collection strategies typically prioritize balancing disease categories while neglecting the importance of subgroup balance based on factors such as age and gender. This imbalance leads to the model's inconsistent performance across subgroups, thereby limiting its clinical applicability. This issue is particularly pronounced in skin cancer diagnosis due to the entanglement and complexity of feature patterns in skin cancer images. In this article, we approach the problem from a new perspective: making the feature patterns extracted by the model more independent and enhancing the representational ability of subgroups. We propose a novel framework that enhances feature independence across channels and improves subgroup representation. Central to our design is the Feature Pattern Separating Module (FPSM), combined with dual subgroupspecific classifiers and an auxiliary subgroup-type predictor. A diverse objective optimization model was introduced to guide the joint optimization of classification accuracy and subgroup fairness. Additionally, we proposed a novel classification loss function to dynamically balance the loss between subgroups. Experiments on two public skin cancer datasets demonstrate that our method improves both overall performance and fairness across subgroups, outperforming existing methods and offering a promising solution for fairer medical AI.

Accurate detection of retinal vessels plays a critical role in reflecting a wide range of health status indicators in the clinical diagnosis of ocular diseases. Recently, advances in deep learning have led to a surge in retinal vessel segmentation methods, which have significantly contributed to the quantitative analysis of vascular morphology. However, retinal vasculature differs significantly from conventional segmentation targets in that it consists of extremely thin and branching structures, whose global morphology varies greatly across images. These characteristics continue to pose challenges to segmentation precision and robustness. To address these issues, we propose MMUNet, a novel architecture tailored for efficient retinal vessel segmentation. The model incorporates Morph Mamba Convolution layers, which replace pointwise convolutions to enhance branching topological perception through morph, state-aware feature sampling. Additionally, Reverse Selective State Guidance modules integrate reverse guidance theory with state-space modeling to improve geometric boundary awareness and decoding efficiency. Extensive experiments conducted on two public retinal vessel segmentation datasets demonstrate the superior performance of the proposed method in segmentation accuracy. Compared to the existing approaches, MM-UNet achieves F1-score gains of 1.64 % on DRIVE and 1.25 % on STARE, demonstrating its effectiveness and advancement. The project code is public via https://github.com/liujiawen-jpg/MM-UNet.

Multi-view clustering leverages complementary information across views but struggles with scalability and robust inter-view fusion. We propose DREAM, a novel binary multi-view clustering method that integrates dual high-order random walks to enhance performance. First, multi-view data are mapped to bipartite graphs via RBF, then refined by high-order random walks to mitigate anchor sensitivity. Projection matrices are regularized by Enhanced Tensor Rank (ETR) for inter-view synergy, Bregman divergence for intra-view specificity, and adaptive weighting for view contribution balancing. Cluster labels are directly derived from the optimized binary codes. Experiments on four benchmarks show DREAM outperforms eight state-of-the-art methods, achieving up to 22.2% higher ACC and 29.7% higher NMI on biological datasets. The source code is available at https://github.com/HsuehBiao/DREAM.

Compared to single-view clustering, multi-view clustering leverages complementary information from different feature representations of the same object, significantly enhancing the understanding of complex data structures and improving robustness. To further address the computational bottleneck caused by high-dimensional and large-scale data, anchor-based multi-view clustering methods have attracted extensive attention in recent years. These methods select a small set of representative anchor points to construct sparse anchor graphs, approximate global similarities with linear complexity, and achieve cross-view alignment through a shared anchor space, balancing both efficiency and performance. This paper proposes an efficient frame-work for large-scale multi-view clustering, termed Consensus-Guided Anchor Graph Alignment for Multi-view Clustering (CAG2). It constructs view-specific local anchor graphs and aligns them with a global anchor graph to preserve cross-view structural consistency, while enforcing anchor similarity constraints to maintain clustering coherence. CAG2integrates three key mechanisms—anchor graph learning, graph alignment, and similarity preservation—into a unified optimization objective, which is efficiently solved via an alternating optimization strategy with theoretical convergence guarantees. Extensive experiments on five benchmark datasets against five state-of-the-art methods demonstrate its robustness and efficiency.

Multimodal representation learning is critical for a wide range of applications, such as multimodal sentiment analysis. Current multimodal representation learning methods mainly focus on the multimodal alignment or fusion strategies, such that the complementary and consistent information among heterogeneous modalities can be fully explored. However, they mistakenly treat the uncertainty noise within each modality as the complementary information, failing to simultaneously leverage both consistent and complementary information while eliminating the aleatoric uncertainty within each modality. To address this issue, we propose a plug-and-play feature causality decomposition method for multimodal representation learning from causality perspective, which can be integrated into existing models with no affects on the original model structures. Specifically, to deal with the heterogeneity and consistency, according to whether it can be aligned with other modalities, the unimodal feature is first disentangled into two parts: modality-invariant (the synergistic information shared by all heterogeneous modalities) and modality-specific part. To deal with complementarity and uncertainty, the modality-specific part is further decomposed into unique and redundant features, where the redundant feature is removed and the unique feature is reserved based on the backdoor-adjustment. The effectiveness of noise removal is supported by causality theory. Finally, the task-related information, including both synergistic and unique components, is further fed to the original fusion module to obtain the final multimodal representations. Extensive experiments show the effectiveness of our proposed strategies.

Hashing technology has exhibited great cross-modal retrieval potential due to its appealing retrieval efficiency and storage effectiveness. Most current supervised cross-modal retrieval methods heavily rely on accurate semantic supervision, which is intractable for annotations with ever-growing sample sizes. By comparison, the existing unsupervised methods rely on accurate sample similarity preservation strategies with intensive computational costs to compensate for the lack of semantic guidance, which causes these methods to lose the power to bridge the semantic gap. Furthermore, both kinds of approaches need to search for the nearest samples among all samples in a large search space, whose process is laborious. To address these issues, this paper proposes an unsupervised dual deep hashing (UDDH) method with semantic-index and content-code for cross-modal retrieval. Deep hashing networks are utilized to extract deep features and jointly encode the dual hashing codes in a collaborative manner with a common semantic index and modality content codes to simultaneously bridge the semantic and heterogeneous gaps for cross-modal retrieval. The dual deep hashing architecture, comprising the head code on semantic index and tail codes on modality content, enhances the efficiency for cross-modal retrieval. A query sample only needs to search for the retrieved samples with the same semantic index, thus greatly shrinking the search space and achieving superior retrieval efficiency. UDDH integrates the learning processes of deep feature extraction, binary optimization, common semantic index, and modality content code within a unified model, allowing for collaborative optimization to enhance the overall performance. Extensive experiments are conducted to demonstrate the retrieval superiority of the proposed approach over the state-of-the-art baselines.

Text data augmentation is an effective strategy for overcoming the challenge of limited sample sizes in many natural language processing (NLP) tasks. This challenge is especially prominent in the few-shot learning (FSL) scenario, where the data in the target domain is generally much scarcer and of lowered quality. A natural and widely used strategy to mitigate such challenges is to perform data augmentation to better capture data invariance and increase the sample size. However, current text data augmentation methods either can’t ensure the correct labeling of the generated data (lacking faithfulness), or can’t ensure sufficient diversity in the generated data (lacking compactness), or both. Inspired by the recent success of large language models (LLM), especially the development of ChatGPT, we propose a text data augmentation approach based on ChatGPT (named ”AugGPT”). AugGPT rephrases each sentence in the training samples into multiple conceptually similar but semantically different samples. The augmented samples can then be used in downstream model training. Experiment results on multiple few-shot learning text classification tasks show the superior performance of the proposed AugGPT approach over state-of-the-art text data augmentation methods in terms of testing accuracy and distribution of the augmented samples.

Understanding the associations among microRNAs (miRNAs), long non-coding RNAs (lncRNAs), and various diseases as biomarkers holds significant biological importance. Developing efficient, straightforward prediction models is essential to reduce the high cost of experimental research. However, most existing methods typically predict miRNA-disease associations (MDAs), lncRNA-disease associations (LDAs), and lncRNA miRNA interactions (LMIs) separately, often relying on both similarities and associations as inputs. These approaches complicate their application across diverse biological and medical domains. Moreover, few models are capable of simultaneously predicting all three types of associations in a unified framework. In this work, we propose a novel and simplified model, called Simplified input strategy for Multiple Associations Prediction (SimpleMAP). Unlike previous approaches, SimpleMAP eliminates the need for similarity networks or external biological data and instead uses only known associations as input, reducing feature contamination and ensuring better generalization. SimpleMAP is designed to predict MDAs, LDAs, and LMIs concurrently, by constructing a three-layer heterogeneous biomolecular network that captures the associations among miRNAs, lncRNAs, and diseases. Our method employs a single, end-to-end architecture based on stacked deep matrix factorization (SDMF) to process sparse input data and learn latent features effectively. SimpleMAP is designed to concurrently predict MDAs, LDAs, and LMIs by constructing a three-layer heterogeneous biomolecular network that captures multi-relational associations among miRNAs, lncRNAs, and diseases. To enhance predictive performance, we incorporate multiple feature integration strategies to fuse representations extracted by SDMF. This streamlined design makes SimpleMAP one of the first models to predict multiple bio-entity associations jointly using only minimal input data, offering a highly scalable and biologically meaningful solution. SimpleMAP demonstrates superior performance against strong baselines. Further validation on two additional datasets involving miRNA-circRNA-disease associations confirms the models robustness and adaptability. Finally, biologically validated case studies underscore the realworld applicability of SimpleMAP for biomarker discovery in complex biological systems. Overall, SimpleMAP introduces a new paradigm in bio-entity association predictionłachieving multi-type, high-performance prediction with minimal input complexityłmaking it a valuable tool for computational biology and biomedical research.

Emerging researchindicates that the degenerative biomarkers associated with Alzheimer's disease (AD) exhibit a non-random distribution within the cerebral cortex, instead following the structural brain network. The alterations in brain networks occur much earlier than the onset of clinical symptoms, thereby affecting the progression of brain disease. In this context, the utilization of computational methods to ascertain the propagation patterns of neuropathological events would contribute to the comprehension of the pathophysiological mechanism involved in the evolution of AD. Despite the encouraging findings achieved by existing graph-based deep learning approaches in analyzing irregular graph data, their applications in identifying the spreading pathway of neuropathology are limited due to two disadvantages. They include (1) lack of a common brain network as an unbiased reference basis for group comparison, and (2) lack of an appropriate mechanism for the identification of propagation patterns. To this end, we propose a proof-of-concept harmonic wavelet neural network (HWNN) to predict the early stage of AD and localize disease-related significant wavelets, which can be used to characterize the spreading pathways of neuropathological events across the brain network. The extensive experiments constructed on both synthetic and real datasets demonstrate that our proposed method achieves superior performance in classification accuracy and statistical power of identifying propagation patterns, compared with other representative approaches.

Spatio-temporal time series forecasting has attracted great attention in various fields, including climate, power, and traffic forecasting. Recently, Spatio-temporal Graph Neural Networks (STGNNs) have shown promising performance in modeling spatial dependencies based on graph neural networks (GNNs) and temporal dependencies based on temporal learning modules. However, most STGNNs do not effectively integrate explicit and implicit relationships between nodes, nor do they adequately capture long and short-term time dependencies. To address these challenges, this paper presents a Quaternion Spatio-temporal Graph Neural Network(QSTGNN). Specifically, the quaternion spatio-temporal graph is constructed firstly, such that the information of both short and long-term time steps are preserved in quaternion feature tensor, and information of multiple explicit graphs and implicit graph are integrated in quaternion graph adjacency matrix. Then, two modules are designed: a 1D quaternion convolution module and a quaternion graph convolution module. In the 1D quaternion convolution module, complex temporal correlations among short and long-term time steps can be well exploited by 1D quaternion convolution operator based on the quaternion Hamilton product. In the quaternion graph convolution module, quaternion graph convolution is designed to characterize nonlinear dependencies among multiple spatial graphs, including explicit and implicit graphs. Extensive experiments are conducted on six datasets, and the results show that QSTGNN achieves state-of-the-art performances over the existing ten methods. Explainable analysis presents that multiple spatial correlations can accurately illustrate the traffic flow and road functional information in real traffic roads.

Mounting evidence has revealed that functional brain networks are intrinsically dynamic, undergoing changes over time, even in the resting-state environment. Notably, recent studies have highlighted the existence of a small number of critical brain regions within each functional brain network that exhibit a flexible role in adapting the geometric pattern of brain connectivity over time, referred to as "temporal hub" regions. Therefore, the identification of these temporal hubs becomes pivotal for comprehending the mechanisms that underlie the dynamic evolution of brain connectivity. However, existing spatio-temporal hub identification methods rely on static network-based approaches, wherein each temporal hub region is independently inferred from individual time-segmented networks without considering their temporal consistency and consequently fails to align the evolution of hubs with the dynamic changes in brain states. To address this limitation, we propose a novel spatio-temporal hub identification method that fully leverages dynamic graph embedding to distinguish temporal hubs from peripheral nodes, in which dynamic graph embeddings are learned from both spatial and temporal dimensions. Specifically, to preserve the temporal consistency of evolving networks, we model the dynamic graph embedding as a physical model of time, where the network-to-network transition is mathematically expressed as a total variation of dynamic graph embedding with respect to time. Furthermore, a Grassmannian manifold optimization scheme is introduced to enhance graph embedding learning and capture the time-varying topology of brain networks. Experimental results on both synthetic and real fMRI data demonstrate superior temporal consistency in hub identification, surpassing conventional approaches.

Traditional clustering methods rely on pairwise affinity to divide samples into different subgroups. However, high-dimensional small-sample (HDLSS) data are affected by the concentration effects, rendering traditional pairwise metrics unable to accurately describe relationships between samples, leading to suboptimal clustering results. This article advances the proposition of employing high-order affinities to characterize multiple sample relationships as a strategic means to circumnavigate the concentration effects. We establish a nexus between different order affinities by constructing specialized decomposable high-order affinities, thereby formulating a uniform mathematical framework. Building upon this insight, a novel clustering method named uniform tensor clustering (UTC) is proposed, which learns a consensus low-dimensional embedding for clustering by the synergistic exploitation of multiple-order affinities. Extensive experiments on synthetic and real-world datasets demonstrate two findings: 1) high-order affinities are better suited for characterizing sample relationships in complex data and 2) reasonable use of different order affinities can enhance clustering effectiveness, especially in handling high-dimensional data.

Tensor spectral clustering (TSC) is a recently proposed approach to robustly group data into underlying clusters. Unlike the traditional spectral clustering (SC), which merely uses pairwise similarities of data in an affinity matrix, TSC aims at exploring their multiwise similarities in an affinity tensor to achieve better performance. However, the performance of TSC highly relies on the design of multiwise similarities, and it remains unclear especially for high-dimension-low-sample-size (HDLSS) data. To this end, this article has proposed a discriminating TSC (DTSC) for HDLSS data. Specifically, DTSC uses the proposed discriminating affinity tensor that encodes the pair-to-pair similarities, which are particularly constructed by the anchor-based distance. HDLSS asymptotic analysis shows that the proposed affinity tensor can explicitly differentiate samples from different clusters when the feature dimension is large. This theoretical property allows DTSC to improve the clustering performance on HDLSS data. Experimental results on synthetic and benchmark datasets demonstrate the effectiveness and robustness of the proposed method in comparison to several baseline methods.

Multiview subspace clustering has shown promising performance in multimedia and data mining applications. However, its employment in large-scale datasets is limited due to its quadratic or even cubic computational complexity. The anchor graph strategy, which selects a few important samples (anchors) to represent the whole data for different views, has been introduced to address this challenge. These methods rely on a heuristic assumption that the correspondence and class structures between the sets of anchors across different views are the same. This assumption ignores the difference in the ordering of anchors with respect to their associated classes and the number of anchors belonging to the same class from different views. As a result, this can lead to unsatisfactory clustering results due to incorrect anchorwise and classwise alignments. To tackle this issue, this article proposes an anchor-based multiview subspace clustering with anchorwise and classwise alignments (AMCA2) method. Specifically, the proposed method simultaneously aligns and fuses multiple anchor graphs anchor wisely and class wisely via learning permutation matrices and utilizing the Hadamard product. To further enhance the clustering performance of AMCA2, we propose a novel anchor selection method called kernel anchor selection (KAS) to select more representative anchors. Extensive experiments on ten benchmark datasets are conducted to show the superiority and effectiveness of AMCA2over the state-of-the-art methods.

Mild cognitive impairment (MCI) represents an early stage of Alzheimer's disease (AD), characterized by subtle clinical symptoms that pose challenges for accurate diagnosis. The quest for the identification of MCI individuals has highlighted the importance of comprehending the underlying mechanisms of disease causation. Integrated analysis of brain imaging and genomics offers a promising avenue for predicting MCI risk before clinical symptom onset. However, most existing methods face challenges in: 1) mining the brain network-specific topological structure and addressing the single nucleotide polymorphisms (SNPs)-related noise contamination and 2) extracting the discriminative properties of brain imaging genomics, resulting in limited accuracy for MCI diagnosis. To this end, a modality-aware discriminative fusion network (MA-DFN) is proposed to integrate the complementary information from brain imaging genomics to diagnose MCI. Specifically, we first design two modality-specific feature extraction modules: the graph convolutional network with edge-augmented self-attention module (GCN-EASA) and the deep adversarial denoising autoencoder module (DAD-AE), to capture the topological structure of brain networks and the intrinsic distribution of SNPs. Subsequently, a discriminative-enhanced fusion network with correlation regularization module (DFN-CorrReg) is employed to enhance inter-modal consistency and between-class discrimination in brain imaging and genomics. Compared to other state-of-the-art approaches, MA-DFN not only exhibits superior performance in stratifying cognitive normal (CN) and MCI individuals but also identifies disease-related brain regions and risk SNPs locus, which hold potential as putative biomarkers for MCI diagnosis.

Although pre-trained large-scale generative models StyleGAN series have proven to be effective in various editing and translation tasks, they are limited to pre-defined fixed aspect ratio. To overcome this limitation, we propose StyleGAN-$\infty$∞, a model that enables pre-trained StyleGAN to perform arbitrary-ratio conditional synthesis. Our key insight is to distill the expressive StyleGAN features into a StyleBook, such that an arbitrary-ratio condition can be translated to other forms by properly assembling pre-defined StyleBook vectors. To learn and leverage the StyleBook, we employ a network with three distinct stages, each corresponding to StyleBook extraction, StyleBook correspondence learning, and arbitrary-ratio synthesis. Extensive experiments on various conditional synthesis tasks, like super-resolution, sketch synthesis, and semantic synthesis, demonstrate superior performances over state-of-the-art image-to-image translation methods. Moreover, our model can easily generate megapixel images in diverse modalities by taking advantage of different pre-trained StyleGAN models.

Recently, single-image SVBRDF capture is formulated as a regression problem, which uses a network to infer four SVBRDF maps from a flash-lit image. However, the accuracy is still not satisfactory since previous approaches usually adopt end-to-end inference strategies. To mitigate the challenge, we propose "auxiliary renderings" as the intermediate regression targets, through which we divide the original end-to-end regression task into several easier sub-tasks, thus achieving better inference accuracy. Our contributions are threefold. First, we design three (or two pairs of) auxiliary renderings and summarize the motivations behind the designs. By our design, the auxiliary images are bumpiness-flattened or highlight-removed, containing disentangled visual cues about the final SVBRDF maps and can be easily transformed to the final maps. Second, to help estimate the auxiliary targets from the input image, we propose two mask images including a bumpiness mask and a highlight mask. Our method thus first infers mask images, then with the help of the mask images infers auxiliary renderings, and finally transforms the auxiliary images to SVBRDF maps. Third, we propose backbone UNets to infer mask images, and gated deformable UNets for estimating auxiliary targets. Thanks to the well-designed networks and intermediate images, our method outputs better SVBRDF maps than previous approaches, validated by the extensive comparisonal and ablation experiments.

With the development of high-throughput technologies, a massive scale of pharmacological and genomic data has been accumulated, which enables the discovery of the correlation between oncogenic genes and therapeutic drugs. Generally, genes with similar functions tend to be related to similar drugs and vice versa. Previous methods detect such associations between gene modules and drug modules based on the similarity of individual genes and drugs, resulting in an inaccurate capture of module-level interactions. How to fully leverage the underlying modules within genes and drugs is a key challenge when identifying regulatory relationships between gene and drug modules. In this paper, we propose a module-level gene-drug interaction identification model via hierarchical optimal transport (H-OT). Particularly, prior knowledge of genes or drugs is integrated to uncover the underlying module of genes or drugs with similar biological functions. Moreover, the optimal associations between gene and drug modules are determined by minimizing high-level OT distance between them, the cost function specified in high-level OT is automatically learned by low-level OT, which incorporates module patterns within genes and drugs. Experiments conducted on synthetic datasets demonstrate that our model exhibits superior performance than six state-of-the-art methods. Additionally, our evaluation of real drug-gene data highlights the model’s statistical power. The gene-drug modules identified by our approach reveal closely related gene-drug interactions and significantly enrich pathways associated with cancer.

Nasopharyngeal carcinoma (NPC) is a malignant tumor whose accurate segmentation is a prerequisite for patient treatment and prognosis. Location and area of lymph nodes (LN) provide significant information for patient staging. However, existing methods for segmenting primary NPC tumors and LNs ignore three major challenges, i.e., uncertain position, irregular boundary, and false negative, thus, obtaining unsatisfactory segmentation results. Given the remarkable success of capturing global context information in Transformer, we have tailored a Transformer-based architecture, named NPCTrans, to address these limitations. NPCTrans is underpinned by three critical modules, including the position locating module (PLM), the boundary attention module (BAM), and the lesion region correction module (LRC). The PLM aims to improve sensitivity towards lesion regions by utilizing relative position encodings and control position information by gated in global attention. The BAM captures and refines the key points of the irregular boundary to enhance boundary segmentation performance. The LRC focuses on multi-scale feature learning by augmenting features of pixels and regions of interest at different scales to correctly identify multiple lesion regions to reduce false negatives. Extensive benchmarking experiments are conducted on our benchmarking dataset, which includes 9,100 samples collected from 723 patients, and results demonstrate that NPCTrans outperforms existing state-of-the-art models.

The Segment Anything Model (SAM) has gained popularity as a versatile image segmentation method, thanks to its strong generalization capabilities across various domains. However, when applied to optic disc (OD) and optic cup (OC) segmentation tasks, SAM encounters challenges due to the complex structures, low contrast, and blurred boundaries typical of fundus images, leading to suboptimal performance. To over-come these challenges, we introduce a novel model, FunduSAM, which incorporates several Adapters into SAM to create a deep network specifically designed for OD and OC segmentation. The FunduSAM utilizes Adapter into each transformer block after encoder for parameter fine-tuning (PEFT). It enhances SAM’s feature extraction capabilities by designing a Convolutional Block Attention Module (CBAM), addressing issues related to blurred boundaries and low contrast. Given the unique requirements of OD and OC segmentation, polar transformation is used to convert the original fundus OD images into a format better suited for training and evaluating FunduSAM. A joint loss is used to achieve structure preservation between the OD and OC, while accurate segmentation. Extensive experiments on the REFUGE dataset, comprising 1,200 fundus images, demonstrate the superior performance of FunduSAM compared to five mainstream approaches.

MOTIVATION: Neoantigens, derived from somatic mutations in cancer cells, can elicit anti-tumor immune responses when presented to autologous T cells by human leukocyte antigen. Identifying immunogenic neoantigens is crucial for cancer immunotherapy development. However, the accuracy of current bioinformatic methods remains unsatisfactory. Surface and structural features of peptide-HLA class I (pHLA-I) complexes offer valuable insight into the immunogenicity of neoantigens. RESULTS: We present NeoaPred, a deep-learning framework for neoantigen prediction. NeoaPred accurately constructs pHLA-I complex structures, with 82.37% of the predicted structures showing an RMSD of < 1 Å. Using these structures, NeoaPred integrates differences in surface, structural, and atom group features between the mutant peptide and its wild-type counterpart to predict a foreignness score. This foreignness score is an effective factor for neoantigen prediction, achieving an AUROC (Area Under the Receiver Operating Characteristic Curve) of 0.81 and an AUPRC (Area Under the Precision-Recall Curve) of 0.54 in the test set, outperforming existing methods. AVAILABILITY AND IMPLEMENTATION: The source code is released under an Apache v2.0 license and is available at the GitHub repository (https://github.com/Dulab2020/NeoaPred).

This paper addresses the problem of animating a person in static images, the core task of which is to infer future poses for the person. Existing approaches predict future poses in the 2D space, suffering from entanglement of pose action and shape. We propose a method that generates actions in the 3D space and then transfers them to the 2D person. We first lift the 2D pose of the person to a 3D skeleton, then propose a 3D action synthesis network predicting future skeletons, and finally devise a self-supervised action transfer network that transfers the actions of 3D skeletons to the 2D person. Actions generated in the 3D space look plausible and vivid. More importantly, self-supervised action transfer allows our method to be trained only on a 3D MoCap dataset while being able to process images in different domains. Experiments on three image datasets validate the effectiveness of our method.

Abstract The traditional microfacet rendering models usually only consider the straight propagation of light and do not take into account the diffraction effect when calculating the radiance of outgoing light. However, ignoring the energy generated by diffraction can lead to darker rendering results when the object's surface has many small details. To address this issue, we introduce a diffraction energy term in the microfacet model to compensate for the energy loss caused by diffraction. Starting from the Fresnel‐Kirchhoff diffraction theorem, we combine it with the Cook‐Torrance model. By incorporating the computed diffraction radiance into the outgoing radiance of the microfacet, we obtain a diffraction‐compensated BRDF (Bidirectional Reflectance Distribution Function) model. Experimental results demonstrate that our proposed method has a significant effect in compensating for outgoing light and produces more realistic rendering results.

Histopathology image-based survival prediction aims to provide a precise assessment of cancer prognosis and can inform personalized treatment decision-making in order to improve patient outcomes. However, existing methods cannot automatically model the complex correlations between numerous morphologically diverse patches in each whole slide image (WSI), thereby preventing them from achieving a more profound understanding and inference of the patient status. To address this, here we propose a novel deep learning framework, termed dual-stream multi-dependency graph neural network (DM-GNN), to enable precise cancer patient survival analysis. Specifically, DM-GNN is structured with the feature updating and global analysis branches to better model each WSI as two graphs based on morphological affinity and global co-activating dependencies. As these two dependencies depict each WSI from distinct but complementary perspectives, the two designed branches of DM-GNN can jointly achieve the multi-view modeling of complex correlations between the patches. Moreover, DM-GNN is also capable of boosting the utilization of dependency information during graph construction by introducing the affinity-guided attention recalibration module as the readout function. This novel module offers increased robustness against feature perturbation, thereby ensuring more reliable and stable predictions. Extensive benchmarking experiments on five TCGA datasets demonstrate that DM-GNN outperforms other state-of-the-art methods and offers interpretable prediction insights based on the morphological depiction of high-attention patches. Overall, DM-GNN represents a powerful and auxiliary tool for personalized cancer prognosis from histopathology images and has great potential to assist clinicians in making personalized treatment decisions and improving patient outcomes.

Tensor spectral clustering (TSC) is an emerging approach that explores multi-wise similarities to boost learning. However, two key challenges have yet to be well addressed in the existing TSC methods: (1) The construction and storage of high-order affinity tensors to encode the multi-wise similarities are memory-intensive and hampers their applicability, and (2) they mostly employ a two-stage approach that integrates multiple affinity tensors of different orders to learn a consensus tensor spectral embedding, thus often leading to a suboptimal clustering result. To this end, this paper proposes a tensor spectral clustering network (TSC-Net) to achieve one-stage learning of a consensus tensor spectral embedding, while reducing the memory cost. TSC-Net employs a deep neural network that learns to map the input samples to the consensus tensor spectral embedding, guided by a TSC objective with multiple affinity tensors. It uses stochastic optimization to calculate a small part of the affinity tensors, thereby avoiding loading the whole affinity tensors for computation, thus significantly reducing the memory cost. Through using an ensemble of multiple affinity tensors, the TSC can dramatically improve clustering performance. Empirical studies on benchmark datasets demonstrate that TSC-Net outperforms the recent baseline methods.

In multi-view environment, it would yield missing observations due to the limitation of the observation process. The most current representation learning methods struggle to explore complete information by lacking either cross-generative via simply filling in missing view data, or solidative via inferring a consistent representation among the existing views. To address this problem, we propose a deep generative model to learn a complete generative latent representation, namely Complete Multi-view Variational Auto-Encoders (CMVAE), which models the generation of the multiple views from a complete latent variable represented by a mixture of Gaussian distributions. Thus, the missing view can be fully characterized by the latent variables and is resolved by estimating its posterior distribution. Accordingly, a novel variational lower bound is introduced to integrate view-invariant information into posterior inference to enhance the solidative of the learned latent representation. The intrinsic correlations between views are mined to seek cross-view generality, and information leading to missing views is fused by view weights to reach solidity. Benchmark experimental results in clustering, classification, and cross-view image generation tasks demonstrate the superiority of CMVAE, while time complexity and parameter sensitivity analyses illustrate the efficiency and robustness. Additionally, application to bioinformatics data exemplifies its practical significance.

Graph-based multi-view clustering encodes multi-view data into sample affinities to find consensus representation, effectively overcoming heterogeneity across different views. However, traditional affinity measures tend to collapse as the feature dimension expands, posing challenges in estimating a unified alignment that reveals both cross-view and inner relationships. To tackle this challenge, we propose to achieve multi-view uniform clustering via consensus representation co-regularization. First, the sample affinities are encoded by both popular dyadic affinity and recent high-order affinities to comprehensively characterize spatial distributions of the HDLSS data. Second, a fused consensus representation is learned through aligning the multi-view low-dimensional representation by co-regularization. The learning of the fused representation is modeled by a high-order eigenvalue problem within manifold space to preserve the intrinsic connections and complementary correlations of original data. A numerical scheme via manifold minimization is designed to solve the high-order eigenvalue problem efficaciously. Experiments on eight HDLSS datasets demonstrate the effectiveness of our proposed method in comparison with the recent thirteen benchmark methods.

3D neural rendering enables photo-realistic reconstruction of a specific scene by encoding discontinuous inputs into a neural representation. Despite the remarkable rendering results, the storage of network parameters is not transmission-friendly and not extendable to metaverse applications. In this paper, we propose an invertible neural rendering approach that enables generating an interactive 3D model from a single image (i.e., 3D Snapshot). Our idea is to distill a pre-trained neural rendering model (e.g., NeRF) into a visualizable image form that can then be easily inverted back to a neural network. To this end, we first present a neural image distillation method to optimize three neural planes for representing the original neural rendering model. However, this representation is noisy and visually meaningless. We thus propose a dynamic invertible neural network to embed this noisy representation into a plausible image representation of the scene. We demonstrate promising reconstruction quality quantitatively and qualitatively, by comparing to the original neural rendering model, as well as video-based invertible methods. On the other hand, our method can store dozens of NeRFs with a compact restoration network (5 MB), and embedding each 3D scene takes up only 160 KB of storage. More importantly, our approach is the first solution that allows embedding a neural rendering model into image representations, which enables applications like creating an interactive 3D model from a printed image in the metaverse.

Increasing evidence has indicated that RNA-binding proteins (RBPs) play an essential role in mediating alternative splicing (AS) events during epithelial-mesenchymal transition (EMT). However, due to the substantial cost and complexity of biological experiments, how AS events are regulated and influenced remains largely unknown. Thus, it is important to construct effective models for inferring hidden RBP-AS event associations during EMT process. In this paper, a novel and efficient model was developed to identify AS event-related candidate RBPs based on Adaptive Graph-based Multi-Label learning (AGML). In particular, we propose to adaptively learn a new affinity graph to capture the intrinsic structure of data for both RBPs and AS events. Multi-view similarity matrices are employed for maintaining the intrinsic structure and guiding the adaptive graph learning. We then simultaneously update the RBP and AS event associations that are predicted from both spaces by applying multi-label learning. The experimental results have shown that our AGML achieved AUC values of 0.9521 and 0.9873 by 5-fold and leave-one-out cross-validations, respectively, indicating the superiority and effectiveness of our proposed model. Furthermore, AGML can serve as an efficient and reliable tool for uncovering novel AS events-associated RBPs and is applicable for predicting the associations between other biological entities.

Autism spectrum disorder (ASD) is accompanied by impaired social cognition and behavior. The expense of supporting patients with ASD turns into a significant problem for society. Parsing neurobiological subtypes is a crucial way for delineating the heterogeneity in autistic brains, with significant implications for improving ASD diagnosis and promoting the development of personalized intervention models. Nevertheless, a comprehensive understanding of the heterogeneity in cortical morphology of ASD is still lacking, and the question of whether neuroanatomical subtypes remain stable during cortical development remains unclear. Here, we used T1-weighted images of 515 male patients with ASD, including 216 autistic children (6–11 years), 187 adolescents (12–17 years), and 112 young adults (18–29 years), along with 595 age and gender-matched typically developing (TD) individuals. Cortical thickness (CT), surface area (SA), and volumes of cortical (CV) and subcortical (SV) regions were extracted. A single network layer was established by calculating the covariance of each feature across brain regions between participants, thereby constructing a multilayer intersubject covariance network. Applying a community detection algorithm to multilayer networks derived from different feature combinations, we observed that the network comprising CT and CV layers exhibited the most prominent modular organization, resulting in three subtypes of ASD for each of the three age groups. Subtypes within the corresponding age group significantly differed in terms of brain morphology and clinical scales. Furthermore, the subtypes of children with ASD underwent reorganization with development, transitioning from childhood to adolescence and adulthood, rather than consistently persist. Additionally, subtype categorization largely improved the diagnostic accuracy of ASD compared to diagnosing the entire ASD cohort. These findings demonstrated distinct neuroanatomical manifestations of ASD subtypes across various developmental periods, highlighting the significance of age-related subtyping in facilitating the etiology and diagnosis of ASD.

Cancer is one of the most challenging health problems worldwide. Accurate cancer survival prediction is vital for clinical decision making. Many deep learning methods have been proposed to understand the association between patients’ genomic features and survival time. In most cases, the gene expression matrix is fed directly to the deep learning model. However, this approach completely ignores the interactions between biomolecules, and the resulting models can only learn the expression levels of genes to predict patient survival. In essence, the interaction between biomolecules is the key to determining the direction and function of biological processes. Proteins are the building blocks and principal undertakings of life activities, and as such, their complex interaction network is potentially informative for deep learning methods. Therefore, a more reliable approach is to have the neural network learn both gene expression data and protein interaction networks. We propose a new computational approach, termed CRESCENT, which is a protein-protein interaction (PPI) prior knowledge graph-based convolutional neural network (GCN) to improve cancer survival prediction. CRESCENT relies on the gene expression networks rather than gene expression levels to predict patient survival. The performance of CRESCENT is evaluated on a large-scale pan-cancer dataset consisting of 5991 patients from 16 different types of cancers. Extensive benchmarking experiments demonstrate that our proposed method is competitive in terms of the evaluation metric of the time-dependent concordance index(${C}^{td}$) when compared with several existing state-of-the-art approaches. Experiments also show that incorporating the network structure between genomic features effectively improves cancer survival prediction.

Spectral clustering with graph learning usually performs eigen-decomposition on the adaptive graph to obtain embedded representation for clustering. In terms of adaptive graph learning, the embedded representation is usually treated as the principal component of the graph to help improve graph structure. However, most adaptive graph learning methods only use a single graph layer. Therefore, the extraction power of embedded representation is restricted to single graph layer and insufficient to explore the intrinsic information. To break through this limitation, this article proposes a stacked network to realize spectral clustering with adaptive graph learning (SCnet-AGL). Specifically, the network allows the development of latent embedded representation underlying the multiple graph layers to reveal the intrinsic information. Meanwhile, we have designed an adaptive graph learning scheme to exploit the latent embedded representation for graph learning. With the advantage of the network, an augmented graph is obtained by incorporating the representation information for graph learning layer by layer. Finally, an efficient algorithm with feedback training scheme is proposed for network training. Experiments on real datasets demonstrate the effectiveness of the proposed network, and show that it is feasible to develop latent embedded representation to improve clustering performance.

Current cross-modal retrieval methods heavily rely on accurate semantic labels or sample similarity measurements, and need to search for the nearest samples among all samples in the huge search space, severely limiting the application in stratifying large-scale and high-dimensional multimodal data. To tackle with the issues, this paper proposes an unsupervised cross-modal retrieval method to bypass the semanticwise supervision and samplewise similarity from a standpoint of featurewise matching, named by unsupervised dual hashing coding (UDC). It jointly learns the dual hashing codes on semantic tagging and sample content through factorizing a feature matching potential, which is allowed to bridge the semantic and heterogeneous gaps among different modalities simultaneously through maintaining the inter-modality-consistent semantic information and cross-modality-correlated sample content. In this way, each sample is uniquely coded by a head code on semanticwise tags, and tail codes on samplewise content. The dual coding design makes it very efficient for sample retrieval, in which the query sample only need to search for the retrieved ones with the same semantic tag, greatly narrowing down the search space. The proposed model avoids the calculation of massive sample-wise similarity and works with dual hashing coding scheme, which achieves a twofold efficiency enhancement for analyzing the large-scale and high-dimensional multimodal data. Extensive experiments have been conducted to demonstrate that it achieved superiority on computational time and retrieval performance.

Mounting evidence shows that Alzheimer's disease (AD) manifests the dysfunction of the brain network much earlier before the onset of clinical symptoms, making its early diagnosis possible. Current brain network analyses treat high-dimensional network data as a regular matrix or vector, which destroys the essential network topology, thereby seriously affecting diagnosis accuracy. In this context, harmonic waves provide a solid theoretical background for exploring brain network topology. However, the harmonic waves are originally intended to discover neurological disease propagation patterns in the brain, which makes it difficult to accommodate brain disease diagnosis with high heterogeneity. To address this challenge, this article proposes a network manifold harmonic discriminant analysis (MHDA) method for accurately detecting AD. Each brain network is regarded as an instance drawn on a Stiefel manifold. Every instance is represented by a set of orthonormal eigenvectors (i.e., harmonic waves) derived from its Laplacian matrix, which fully respects the topological structure of the brain network. An MHDA method within the Stiefel space is proposed to identify the group-dependent common harmonic waves, which can be used as group-specific references for downstream analyses. Extensive experiments are conducted to demonstrate the effectiveness of the proposed method in stratifying cognitively normal (CN) controls, mild cognitive impairment (MCI), and AD.

Reduced-order simulation is an emerging method for accelerating physical simulations with high DOFs, and recently developed neural-network-based methods with nonlinear subspaces have been proven effective in diverse applications as more concise subspaces can be detected. However, the complexity and landscape of simulation objectives within the subspace have not been optimized, which leaves room for enhancement of the convergence speed. This work focuses on this point by proposing a general method for finding optimized subspace mappings, enabling further acceleration of neural reduced-order simulations while capturing comprehensive representations of the configuration manifolds. We achieve this by optimizing the Lipschitz energy of the elasticity term in the simulation objective, and incorporating the cubature approximation into the training process to manage the high memory and time demands associated with optimizing the newly introduced energy. Our method is versatile and applicable to both supervised and unsupervised settings for optimizing the parameterizations of the configuration manifolds. We demonstrate the effectiveness of our approach through general cases in both quasi-static and dynamics simulations. Our method achieves acceleration factors of up to 6.83 while consistently preserving comparable simulation accuracy in various cases, including large twisting, bending, and rotational deformations with collision handling. This novel approach offers significant potential for accelerating physical simulations, and can be a good add-on to existing neural-network-based solutions in modeling complex deformable objects.

The knowledge graph (KG) is a highly needed basis to support the high-fidelity and high-interpretability modeling of various tasks in healthcare artificial intelligence. In this work, we focus on constructing an oncology knowledge graph that will be used in downstream cancer research and solution development. Modern supervised learning for knowledge graph construction requires a large amount of manually labeled data, which makes the process time-consuming and labor-intensive. Although there exists multiple research on named entity recognition and relation extraction based on distantly supervised learning, constructing a domain-specific knowledge graph from large collections of textual data without manual annotations is still an urgent problem to be solved. In response, we propose an integrated framework for adapting and re-learning knowledge graphs from a general domain (biomedical in our case) to a fine-defined domain (oncology). In this framework, we apply distant-supervision on cross-domain knowledge graph adaptation. Consequently, no manual data annotation is required to train the model. We introduce a novel iterative training strategy to facilitate the discovery of domain-specific named entities and triplets. Experimental results indicate that the proposed framework can perform domain adaptation and construction of knowledge graphs efficiently.

Due to the inherent high-dimensional characteristics of genomic data, traditional single metric/kernel-based clustering methods fail to accurately perform data analysis. To address this issue, we propose a multi-kernel clustering with tensor fusion on the Grassmann manifold (MKCTM). Specifically, multiple kernel functions are employed to map data into different kernel spaces and utilize tensor representations to capture their high-order relationships. By introducing a tensor low-rank constraint, we maximize the correlation among kernels while separating the noise and redundancy information from kernel tensor. Finally, the learned kernel tensor is fused on the Grassmann manifold to obtain the final kernel matrix for enhancing clustering. We integrate tensor learning and tensor fusion steps into a unified optimization model and propose an efficient iterative optimization algorithm to solve it. Our proposed method is evaluated on six high-dimensional gene expression datasets against eight popular baseline methods. The remarkable experimental performance demonstrates the exceptional effectiveness of our approach. Our code is available at https://github.com/foureverfei/MKCTM.git

Due to domain shift, a large performance drop is usually observed when a trained crowd counting model is deployed in the wild. While existing domain-adaptive crowd counting methods achieve promising results, they typically regard each crowd image as a whole and reduce domain discrepancies in a holistic manner, thus limiting further improvement of domain adaptation performance. To this end, we propose to untangle domain-invariant crowd and domain-specific background from crowd images and design a fine-grained domain adaption method for crowd counting. Specifically, to disentangle crowd from background, we propose to learn crowd segmentation from point-level crowd counting annotations in a weakly-supervised manner. Based on the derived segmentation, we design a crowd-aware domain adaptation mechanism consisting of two crowd-aware adaptation modules, i.e., Crowd Region Transfer (CRT) and Crowd Density Alignment (CDA). The CRT module is designed to guide crowd features transfer across domains beyond background distractions. The CDA module dedicates to regularising target-domain crowd density generation by its own crowd density distribution. Our method outperforms previous approaches consistently in the widely-used adaptation scenarios.

Point clouds obtained by LiDAR and other sensors are usually sparse and irregular. Low-quality point clouds have serious influence on the final performance of downstream tasks. Recently, a point cloud upsampling network with normalizing flows has been proposed to address this problem. However, the network heavily relies on designing specialized architectures to achieve invertibility. In this paper, we propose a novel invertible residual neural network for point cloud upsampling, called PU-INN, which allows unconstrained architectures to learn more expressive feature transformations. Then, we propose a conditional injector to improve nonlinear transformation ability of the neural network while guaranteeing invertibility. Furthermore, a lightweight interpolator is proposed based on semantic similarity distance in the latent space, which can intuitively reflect the interpolation changes in Euclidean space. Qualitative and quantitative results show that our method outperforms the state-of-the-art works in terms of distribution uniformity, proximity-to-surface accuracy, 3D reconstruction quality, and computation efficiency.

As a challenging task, unsupervised person re-identification (Re-ID) aims to optimize the pedestrian matching model based on the unlabeled image frames from surveillance videos. Recently, the fusion with the spatio-temporal clues of pedestrians have been proven effective to improve the performance of classification. However, most of these methods adopt some hard combination approaches by multiplying the visual scores with the spatio-temporal scores, which are sensitive to the noise caused by imprecise estimation of the spatio-temporal patterns in unlabeled datasets and limit the advantage of the fusion model. In this paper, we propose a Graph based Spatio-Temporal Fusion model for high-performance multi-modal person Re-ID, namely G-Fusion, to mitigate the impact of noise. In particular, we construct a graph of pedestrian images by selecting neighboring nodes based on the visual information and the transition time between cameras. Then we use a randomly initialized two-layer GraphSAGE model to obtain the multi-modal affinity matrix between images, and deploy the distillation learning to optimize the visual model by learning the affinity between the nodes. Finally, a graph-based multi-modal re-ranking method is deployed to make the decision in the testing phase for precise person Re-ID. Comprehensive experiments are conducted on two large-scale Re-ID datasets, and the results show that our method achieves a significant improvement of the performance while combined with SOTA unsupervised person Re-ID methods. Specifically, the mAP scores can reach 92.2%, and 80.4% on the Market-1501, and MSMT17 datasets respectively.

In an era of more diverse data modalities, multi-view clustering has become a fundamental tool for comprehensive data analysis and exploration. However, existing multi-view unsupervised learning methods often rely on strict assumptions on semantic consistency among samples. In this paper, we reformulate the multi-view clustering problem from an information-theoretic perspective and propose a general theoretical model. In particular, we define three desiderata under multi-view unsupervised learning in terms of mutual information, namely, comprehensiveness, concentration, and cross-diversity. The multi-view variational lower bound is then obtained by approximating the samples' high-dimensional mutual information. The Kullback–Leibler divergence is utilized to deduce sample assignments. Ultimately the information-based multi-view clustering model leverages deep neural networks and Stochastic Gradient Variational Bayes to achieve representation learning and clustering simultaneously. Extensive experiments on both synthetic and real datasets with wide types demonstrate that the proposed method exhibits a more stable and superior clustering performance than state-of-the-art algorithms.

Identifying regulatory modules between miRNAs and genes is crucial in cancer research. It promotes a comprehensive understanding of the molecular mechanisms of cancer. The genomic data collected from subjects usually relate to different cancer statuses, such as different TNM Classifications of Malignant Tumors (TNM) or histological subtypes. Simple integrated analyses generally identify the core of the tumorigenesis (common modules) but miss the subtype-specific regulatory mechanisms (specific modules). In contrast, separate analyses can only report the differences and ignore important common modules. Therefore, there is an urgent need to develop a novel method to jointly analyze miRNA and gene data of different cancer statuses to identify common and specific modules. To that end, we developed a High-Order Graph Matching model to identify Common and Specific modules (HOGMCS) between miRNA and gene data of different cancer statuses. We first demonstrate the superiority of HOGMCS through a comparison with four state-of-the-art techniques using a set of simulated data. Then, we apply HOGMCS on stomach adenocarcinoma data with four TNM stages and two histological types, and breast invasive carcinoma data with four PAM50 subtypes. The experimental results demonstrate that HOGMCS can accurately extract common and subtype-specific miRNA-gene regulatory modules, where many identified miRNA-gene interactions have been confirmed in several public databases.

Identifying cancer subtypes holds essential promise for improving prognosis and personalized treatment. Cancer subtyping based on multi-omics data has become a hotspot in bioinformatics research. One of the critical approaches of handling data heterogeneity in multi-omics data is first modeling each omics data as a separate similarity graph. Then, the information of multiple graphs is integrated into a unified graph. However, a significant challenge is how to measure the similarity of nodes in each graph and preserve cluster information of each graph. To that end, we exploit a new high order proximity in each graph and propose a similarity fusion method to fuse the high order proximity of multiple graphs while preserving cluster information of multiple graphs. Compared with the current techniques employing the first order proximity, exploiting high order proximity contributes to attaining accurate similarity. The proposed similarity fusion method makes full use of the complementary information from multi-omics data. Experiments in six benchmark multi-omics datasets and two individual cancer case studies confirm that our proposed method achieves statistically significant and biologically meaningful cancer subtypes.

DNA sharing and querying of personal genomic sequences have becoming more critical than ever with the accumulation of large-scale biomedical data. Quantitative genomic studies heavily rely on multi-source DNA datasets from different institutions, who are reluctant to share the data via cloud centers. The high sensitivity of DNA has compelled the government to restrict its acquisition and usage. One potential solution to tackle the issue is designing a secure query strategy on the encrypted DNA datasets. However, the popular secure DNA query schemes remain defective in verifiability, reliability, and NDA privacy. To relieve the issues, we propose a DNA searchable encryption method named EncGD to achieve verifiable privacy-preserving queries and reliable updates on multi-source dynamic DNA datasets. Specifically, the proposed scheme EncGD is designed by the plaintext-related permutation and substitution primitives, which can enhance the DNA privacy due to the chosen-plaintext attack (CPA)-resist ability. Furthermore, the method realizes the verifiability and reliability by adding known information to block the malicious behaviors of clouds. Experimental results demonstrate the superior efficiency of the proposed method comparing with the two state-of-the-art schemes in terms of time and space costs.

Unsupervised feature selection is a vital yet challenging topic for effective data learning. Recently, 2-D feature selection methods show good performance on image analysis by utilizing the structure information of image. Current 2-D methods usually adopt a sparse regularization to spotlight the key features. However, such scheme introduces additional hyperparameter needed for pruning, limiting the applicability of unsupervised algorithms. To overcome these challenges, we design a feature filter to estimate the weight of image features for unsupervised feature selection. Theoretical analysis shows that a sparse regularization can be derived from the feature filter by transformation, indicating that the filter plays the same role as the popular sparse regularization does. We deploy two distinct strategies in terms of feature selection, called multiple feature filters and single common feature filter. The former divides the optimization problem into multiple independent subproblems and selects features that meet the respective interests of each subproblem. The latter selects features that are in the interest of the overall optimization problem. Extensive experiments on seven benchmark datasets show that our unsupervised 2-D weight-based feature selection methods achieve superior performance over the state-of-the-art methods.

Unsupervised graph embedding aims to extract highly discriminative node representations that facilitate the subsequent analysis. Converging evidence shows that a multiview graph provides a more comprehensive relationship between nodes than a single-view graph to capture the intrinsic topology. However, little attention has been paid to excavating discriminative representations of each node from multiview heterogeneous networks in an unsupervised manner. To that end, we propose a novel unsupervised multiview graph embedding method, called multiview deep graph infomax (MVDGI). The backbone of our proposed model sought to maximize the mutual information between the view-dependent node representations and the fused unified representation via contrastive learning. Specifically, the MVDGI first uses an encoder to extract view-dependent node representations from each single-view graph. Next, an aggregator is applied to fuse the view-dependent node representations into the view-independent node representations. Finally, a discriminator is adopted to extract highly discriminative representations via contrastive learning. Extensive experiments demonstrate that the MVDGI achieves better performance than the benchmark methods on five real-world datasets, indicating that the obtained node representations by our proposed approach are more discriminative than by its competitors for classification and clustering tasks.

Since brain network organization is essentially governed by the harmonic waves derived from the Eigen-system of the underlying Laplacian matrix, discovering the harmonic-based alterations provides a new window to understand the pathogenic mechanism of Alzheimer's disease (AD) in a unified reference space. However, current reference (common harmonic waves) estimation studies over the individual harmonic waves are often sensitive to outliers, which are obtained by averaging the heterogenous individual brain networks. To address this challenge, we propose a novel manifold learning approach to identify a set of outlier-immunized common harmonic waves. The backbone of our framework is calculating the geometric median of all individual harmonic waves on the Stiefel manifold, instead of Fréchet mean, thus improving the robustness of learned common harmonic waves to the outliers. A manifold optimization scheme with theoretically guaranteed convergence is tailored to solve our method. The experimental results on synthetic data and real data demonstrate that the common harmonic waves learned by our approach are not only more robust to the outliers than the state-of-the-art methods, but also provide a putative imaging biomarker to predict the early stage of AD.

Numerous studies have shown that accurate analysis of neurological disorders contributes to the early diagnosis of brain disorders and provides a window to diagnose psychiatric disorders due to brain atrophy. The emergence of geometric deep learning approaches provides a new way to characterize geometric variations on brain networks. However, brain network data suffer from high heterogeneity and noise. Consequently, geometric deep learning methods struggle to identify discriminative and clinically meaningful representations from complex brain networks, resulting in poor diagnostic accuracy. Hence, the primary challenge in the diagnosis of brain diseases is to enhance the identification of discriminative features. To this end, this paper presents a dual-attention deep manifold harmonic discrimination (DA-DMHD) method for early diagnosis of neurodegenerative diseases. Here, a low-dimensional manifold projection is first learned to comprehensively exploit the geometric features of the brain network. Further, attention blocks with discrimination are proposed to learn a representation, which facilitates learning of group-dependent discriminant matrices to guide downstream analysis of group-specific references. Our proposed DA-DMHD model is evaluated on two independent datasets, ADNI and ADHD-200. Experimental results demonstrate that the model can tackle the hard-to-capture challenge of heterogeneous brain network topological differences and obtain excellent classifying performance in both accuracy and robustness compared with several existing state-of-the-art methods.

Multi-view clustering aims at integrating information from different views to improve clustering performance. Recent methods integrate multiple view-specific partition matrices to seek a consensus one and have demonstrated promising clustering performance in various applications. However, the clustering performance of such methods heavily relies on the consensus partition matrix estimated by the arithmetic mean in euclidean space and thus is highly susceptible to noise corruption. To this end, this article proposes to learn a consensus partition matrix through the geometric mean on the manifold to achieve robust clustering. Specifically, the multiple view-specific partition matrices can be regarded as points residing in the Stiefel manifold and enable a manifold-based integration. Consequently, the view-specific partition matrices are integrated by estimating a consensus partition matrix as the center point on the Stiefel manifold. Such a partition integration boils down to the Fréchet mean problem on a manifold, which is solved by the intrinsic manifold-based optimization and proves effective in providing a more robust estimation against noise. Experimental results on seven benchmark datasets demonstrate the effectiveness and noise-robustness of our proposed method in comparison to eight competitive methods.

Whole slide image (WSI) analysis represents the current gold standard for cancer diagnosis. To date many fully supervised learning methods have been proposed for WSI classification and segmentation. However, these methods are substantially limited by accurate pixel-level labels, which are labor-intensive to obtain. To solve this problem, we developed an end-to-end multiple instance learning (MIL)-based network for WSI segmentation using coarse-grained labels only. Our network consists of two main components. First, we introduce a hybrid transformer architecture, which uses a fusion mechanism to fuse the feature maps of the convolutional neural network (CNN) and transformer. Second, a novel regional MIL aggregator is proposed, which is used to identify the key instances and address the problem of data imbalance. Unlike the current MIL methods that treat each instance as being independent, our method gathers the information from neighborhood pixels of each instance and captures the correlation between instances. We evaluated our network on CAMELYON16. The benchmarking experiments and ablation studies show that the performance of our method is competitive with those of fully supervised methods and is also better than those of previous MIL segmentation methods.

Accurate nasopharyngeal carcinoma (NPC) segmentation is significant in preventing local recurrence and improving patients’ survival rates. However, existing deep learning-based methods often yield unsatisfactory segmentation results, especially in fine-grained detail. Because NPC is a tiny and infiltrative tumor with a huge background, traditional deep neural networks tend to be dominated by salient information, thus missing the fine-grained details of NPC. To achieve accurate NPC segmentation, a relearning controversial regions method (Reler) is proposed. It consists of three modules, including the controversial features generator (CFG), controversial features finding module (CFF), and controversial regions arbitration module (CRA). First, CFG constructs global and local feature extractors to generate two types of different features. Then, CFF finds the controversial features and corresponding regions by comparing the global and local features’ estimates of the segmentation results of the same input regions. Next, the CRA focuses on controversial features, relearns new features, and produces new segmentation results through a proposed Transformer-based self-attention network. Finally, the uncontroversial segmentation results from CFF and CRA are combined as the final segmentation results. Extensive experiments are conducted on a large NPC dataset containing 6342 images from 596 patients. The experimental results show that the proposed method Reler is effective and superior to the nine state-of-the-art methods.

Multi-view clustering divides data into their under-lying partitions by exploiting multiple views information. Popular approaches leverage cross-view information by self-expressive tensor learning and then learn a low-rank or sparse essential representation tensor for capturing the global structure of multi-view data. However, this process may encounter instability due to the lack of protection for local within-view structures. To overcome this problem, this paper proposes a unified L ow-rank and HyperGraph Laplacian regularized Tensor learning (LHGT) method for multi-view clustering, which aims to integrate within-view high-order affinities in self-expressive tensor learning for capturing inherent clustering structure. LHGT effectively extracts global cross-view and local within-view high-order statistics. An effective optimization procedure is tailored for the proposed model. Experimental results on six real-world datasets illustrate the efficacy of LHGT, where a clear advance over nine state-of-the-art approaches.

Nasopharyngeal carcinoma (NPC) is a malignant tumor whose accurate segmentation is a prerequisite for treatment. However, existing deep learning methods achieve unsatisfactory segmentation performance on NPC MR images, since NPC is infiltrative with small ambiguous boundary volume, making it indiscernible from tightly connected surrounding and complex backgrounds. To address the issues, a NPC segmentation network, termed NPCFormer, is proposed. The NPCFormer consists of two modules, Skip Residual Transformer (SRT) and Boundary Attention Unit (BAU), which are designed for NPC segmentation. The two modules are proposed via redesigning the multi-head self-attention to achieve accurate segmentation of NPC. The SRT exploits the global position context for the NPC locating. The BAU discriminates the tumor boundaries from its surrounding tissues by utilizing the global position context. Extensive experiments on our dataset demonstrate the proposed NPCFormer could distinguish and segment NPC from complex background tissues accurately.

Multi-graph clustering aims at integrating complementary information across multiple graphs to partition multi-view data into underlying clusters. Most current methods rely on pairwise graphs to characterize each view and then employ popular Euclidean averaging to integrate multiple graphs. How-ever, operations of the pairwise graphs on Euclidean space result in insufficient robustness to noise. To address the issue, we propose a method called multi-hypergraph clustering on the Stiefel manifold. First, a hypergraph for each view is constructed to extract high-order relations, which are more resistant to the noise than pairwise graphs. Second, a consensus partition matrix is derived through integrating the multiple hypergraphs on the Stiefel manifold. Such integration is completely driven by the manifold-based operation and enables an effective fusion to mitigate noise contamination, thus improving multi-view clustering performance. Empirical evaluations on five benchmark datasets have demonstrated that our method achieves consistent performance improvement compared with six baseline methods.

The performance of most clustering methods hinges on the used pairwise affinity, which is usually denoted by a similarity matrix. However, the pairwise similarity is notoriously known for its vulnerability of noise contamination or the imbalance in samples or features, and thus hinders accurate clustering. To tackle this issue, we propose to use information among samples to boost the clustering performance. We proved that a simplified similarity for pairs, denoted by a fourth order tensor, equals to the Kronecker product of pairwise similarity matrices under decomposable assumption, or provide complementary information for which the pairwise similarity missed under indecomposable assumption. Then a high order similarity matrix is obtained from the tensor similarity via eigenvalue decomposition. The high order similarity capturing spatial information serves as a robust complement for the pairwise similarity. It is further integrated with the popular pairwise similarity, named by IPS2, to boost the clustering performance. Extensive experiments demonstrated that the proposed IPS2 significantly outperformed previous similarity-based methods on real-world datasets and it was capable of handling the clustering task over under-sampled and noisy datasets.

Multiple modality clustering seeks to partition objects via leveraging cross-modality relations to provide comprehensive descriptions of the same objects. Current clustering methods rely heavily on accurate affinity measurements among samples. The samplewise affinity is costive to be constructed yet easy to corrupt by the heterogeneous gap. In the era of big data, fast and accurate clustering of multiple modality data remains challenging. To fill the gap, we propose a novel approach to achieve the clustering by focusing on feature matching across different modalities instead of samplewise affinity. First, a feature matching matrix is calculated by measuring the potential featurewise correlations. The obtained matching matrix is decomposed into two bases corresponding to the column and row spaces of feature matching, acting as coded bases within feature spaces of the different modalities. Then, the sample assignment is obtained by jointly reconstructing the samples by the two bases. The feature matching potential and sample assignment are collaboratively learned by an alternating optimization scheme. The proposed method dramatically reduces the computational cost by avoiding the costive samplewise affinity estimation, without sacrificing accuracy. Extensive experiments on the synthetic and real-world datasets demonstrate its superior speed and high accuracy.

Multiview clustering seeks to partition objects via leveraging cross-view relations to provide a comprehensive description of the same objects. Most existing methods assume that different views are linear transformable or merely sampling from a common latent space. Such rigid assumptions betray reality, thus leading to unsatisfactory performance. To tackle the issue, we propose to learn both common and specific sampling spaces for each view to fully exploit their collaborative representations. The common space corresponds to the universal self-representation basis for all views, while the specific spaces are the view-specific basis accordingly. An iterative self-supervision scheme is conducted to strengthen the learned affinity matrix. The clustering is modeled by a convex optimization. We first solve its linear formulation by the popular scheme. Then, we employ the deep autoencoder structure to exploit its deep nonlinear formulation. The extensive experimental results on six real-world datasets demonstrate that the proposed model achieves uniform superiority over the benchmark methods.

Identifying gene-drug interactions is vital to understanding biological mechanisms and achieving precise drug repurposing. High-throughput technologies produce a large amount of pharmacological and genomic data, providing an opportunity to explore the associations between oncogenic genes and therapeutic drugs. However, most studies only focus on "one-to-one" or "one-to-many" interactions, ignoring the multivariate patterns between genes and drugs. In this article, a high-order graph matching model with hypergraph constraints is proposed to discover the gene-drug common regulatory modules. Moreover, the prior knowledge is formulated into hypergraph constraints to reveal their multiple correspondences, penalizing the tensor matching process. The experimental results on the synthetic data demonstrate the proposed model is robust to noise contamination and outlier corruption, achieving a better performance than four state-of-the-art methods. We then evaluate the statistical power of our proposed method on the pharmacogenomics data. Our identified gene-drug common modules not only show significantly enriched pathways associated with cancer but also manifest the highly close gene-drug interactions.

Functional connectivities (FC) of brain network manifest remarkable geometric patterns, which is the gateway to understanding brain dynamics. In this work, we present a novel geometric-attention neural network to characterize the time-evolving brain state change from the functional neuroimages by tracking the trajectory of functional dynamics on high-dimension Riemannian manifold of symmetric positive definite (SPD) matrices. Specifically, we put the spotlight on learning the common state-specific manifold signatures that represent the underlying cognition. In this context, the driving force of our neural network is tied up with the learning of the evolution functionals on the Riemannian manifold of SPD matrix that underlies the known evolving brain states. To do so, we train a convolution neural network (CNN) on the Riemannian manifold of SPD matrices to seek for the putative low-dimension feature representations, followed by an end-to-end recurrent neural network (RNN) to yield the time-varying mapping function of SPD matrices which fits the evolutionary trajectories of the underlying states. Furthermore, we devise a geometric attention mechanism in CNN, allowing us to discover the latent geometric patterns in SPD matrices that are associated with the underlying states. Notably, our work has the potential to understand how brain function emerges behavior by investigating the geometrical patterns from functional brain networks, which is essentially a correlation matrix of neuronal activity signals. Our proposed manifold-based neural network achieves promising results in predicting brain state changes on both simulated data and task functional neuroimaging data from Human Connectome Project, which implies great applicability in neuroscience studies.

Nasopharyngeal carcinoma (NPC) is a malignant tumor whose survivability is greatly improved if early diagnosis and timely treatment are provided. Accurate segmentation of both the primary NPC tumors and metastatic lymph nodes (MLNs) is crucial for patient staging and radiotherapy scheduling. However, existing studies mainly focus on the segmentation of primary tumors, eliding the recognition of MLNs, and thus fail to comprehensively provide a landscape for tumor identification. There are three main challenges in segmenting primary NPC tumors and MLNs: variable location, variable size, and irregular boundary. To address these challenges, we propose an automatic segmentation network, named by NPCNet, to achieve segmentation of primary NPC tumors and MLNs simultaneously. Specifically, we design three modules, including position enhancement module (PEM), scale enhancement module (SEM), and boundary enhancement module (BEM), to address the above challenges. First, the PEM enhances the feature representations of the most suspicious regions. Subsequently, the SEM captures multiscale context information and target context information. Finally, the BEM rectifies the unreliable predictions in the segmentation mask. To that end, extensive experiments are conducted on our dataset of 9124 samples collected from 754 patients. Empirical results demonstrate that each module realizes its designed functionalities and is complementary to the others. By incorporating the three proposed modules together, our model achieves state-of-the-art performance compared with nine popular models.

Present studies have discovered that state-of-the-art deep learning models can be attacked by small but well-designed perturbations. Existing attack algorithms for the image captioning task is time-consuming, and their generated adversarial examples cannot transfer well to other models. To generate adversarial examples faster and stronger, we propose to learn the perturbations by a generative model that is governed by three novel loss functions. Image feature distortion loss is designed to maximize the encoded image feature distance between original images and the corresponding adversarial examples at the image domain, and local-global mismatching loss is introduced to separate the mapping encoding representation of the adversarial images and the ground true captions from a local and global perspective in the common semantic space as far as possible cross image and caption domain. Language diversity loss is to make the image captions generated by the adversarial examples as different as possible from the correct image caption at the language domain. Extensive experiments show that our proposed generative model can efficiently generate adversarial examples that successfully generalize to attack image captioning models trained on unseen large-scale datasets or with different architectures, or even the image captioning commercial service.

The topology of human functional networks is assumed to oscillate during brain states changes. The functional neuroimage is employed to offer a non-invasive window to understand cognition and behaviors by characterizing the functional connections between spatially distinct brain regions. Consequently, identifying the transitions of functional connectivities is the critical step to understanding the mechanism of cognition that might be underlined with neurological disorders. However, little attention has been paid to studying the geometry of the entire functional brain network. To tackle this issue, this paper models the cognition changes on functional brain networks as a set of landmarks residing on a Riemannian manifold. Accordingly, we propose a Riemannian manifold mean shift method to detect cognition changes by identifying the representative function networks of the distribution of functional networks. The manifold mean shift (MMS) method is applied on both simulated data and real functional neuroimaging data, downloaded from Human Connectome Project (HCP). Experimental results demonstrated the MMS achieved highly accurate and consistent cognition change, by comparing three state-of-the-art methods.

Accurate and reliable segmentation of nasopharyngeal carcinoma (NPC) in magnetic resonance image (MRI) is important for treatment planning and follow-up evaluation. However, it is still challenging because NPC is infiltrative with a vague border and has a tiny volume with varying sizes and shapes. The above problems easily cause the NPC tumors’ features to submerge in the process of feature extraction. Standard segmentation methods do not cope with the “feature submergence” and performed unsatisfactorily in NPC segmentation. To address the problem, a dual-supervised method equipped with the detection-and-excitation module (DEM) was proposed. DEM strengthens the region of interest (ROI) and weakens complex and immense background through incorporating detection feature maps (outline maps) with intermediated segmentation features. To be specific, the DEM explored the outline of the ROI and output a detection feature map by using a supervised layer guided by a tailored loss function. Then, the DEM uses the results to recalibrates the intermediated segmentation. Finally, a neural network equipped with the proposed DEM was elaborately designed to achieve accurate segmentation. We applied the proposed network termed as DENet on a real nasopharyngeal carcinoma dataset to realize automatic NPC tumors segmentation in multi-modality MR images. The proposed DEM enables the neural network to segment NPC tumors within the immense and complex background. Experimental results have demonstrated the effectiveness of the proposed method by comparing it with benchmark models.

Multi-omics data clustering aims at assigning elements to their respective categories via exploiting the complementary information among multi-omics samples, with the widely application on data mining, bioinformatics, biomedicine and so on. Current multi-omics data clustering methods rely heavily on accurate sample-wise affinity estimation, which is limited by large heterogeneous gap among different sampling spaces. In this paper, we propose a novelly collaborative clustering method, which measures feature-wise matching instead of sample-wise affinity for sample assignment via the guidance of the highly matched features. The method consists of three modules. Firstly, the feature matching is learned by feature-wise correlation. Secondly, the feature matching matrix is factorized into two coded bases within yielded feature spaces. Thirdly, the sample assignment is achieved by joint sample reconstruction employing the two bases. The sample-wise assignment and feature-wise matching are optimized in a collaborative scheme with mutually reinforcing relationship. Extensive experiments demonstrate the superior performance of our proposed method compared with benchmark ones.

Transferring makeup from the misaligned reference image is challenging. Previous methods overcome this barrier by computing pixel-wise correspondences between two images, which is inaccurate and computational-expensive. In this paper, we take a different perspective to break down the makeup transfer problem into a two-step extraction-assignment process. To this end, we propose a Style-based Controllable GAN model that consists of three components, each of which corresponds to target style-code encoding, face identity features extraction, and makeup fusion, respectively. In particular, a Part-specific Style Encoder encodes the component-wise makeup style of the reference image into a style-code in an intermediate latent space W. The style-code discards spatial information and therefore is invariant to spatial misalignment. On the other hand, the style-code embeds component-wise information, enabling flexible partial makeup editing from multiple references. This style-code, together with source identity features, is integrated into a Makeup Fusion Decoder equipped with multiple AdaIN layers to generate the final result. Our proposed method demonstrates great flexibility on makeup transfer by supporting makeup removal, shade-controllable makeup transfer, and part-specific makeup transfer, even with large spatial misalignment. Extensive experiments demonstrate the superiority of our approach over state-of-the-art methods. Code is available at https://github.com/makeuptransfer/SCGAN.

Recently, multi-view clustering methods based on high-order sample affinities to ease learning complex structures attract much attention. However, most of the methods used pre-defined similarity, which is easy to be corrupted by noises and yield suboptimal performance. To tackle with this issue, this paper proposes a novel multi-view clustering method, named by WHSF, which seeks to learn a self-weighted high-order similarity. The high-order similarity is formulated to flexibly capture the intrinsic structure of data, characterized by fusing the interactions across views. A high-order regularization based on the defined similarity is incorporated into the model and assigned with weight parameters, enabling the model to focus on mutual information among views. Extensive experiments on four real-world datasets show that the proposed WHSF outperforms benchmark multi-view methods and can reveal a reliable structure concealed across multiple views.

Clustering objects remains challenges in seeking an under-lying partition by exploiting multiple views. Popular clustering algorithms focus on designing various constraints to handle particular representation tasks, all of which rely on a predefined pairwise similarity (sample-to-sample). However, the pairwise similarity is notoriously vulnerable to noise or outliers contaminations, resulting in sub-optimal clustering performances. To tackle the issue, this paper proposes to enhance multi-view clustering by exploring varieties of high-order statistics within multi-view data, named by HIgh-order Similarity and essential Tensor clustering method (HIST). The HIST incorporates both high-order similarity (samples-to-samples) and high-order correlation (view-to-view) into an adaptive learning model to comprehensively exploit the inherent clustering structure. Experimental results on six real datasets show the superiority of our approach over the ten popular methods.

Accurately identifying the interactions between genomic factors and the response of cancer drugs plays important roles in drug discovery, drug repositioning and cancer treatment. A number of studies revealed that interactions between genes and drugs were 'many-genes-to-many drugs' interactions, i.e. common modules, opposed to 'one-gene-to-one-drug' interactions. Such modules fully explain the interactions between complex biological regulatory mechanisms and cancer drugs. However, strategies for effectively and robustly identifying the underlying common modules among pharmacogenomics data remain to be improved. In this paper, we aim to provide a detailed evaluation of three categories of state-of-the-art common module identification techniques from a machine learning perspective, including non-negative matrix factorization (NMF), partial least squares (PLS) and network analyses. We first evaluate the performance of six methods, namely SNMNMF, NetNMF, SNPLS, O2PLS, NSBM and HOGMMNC, using two series of simulated data sets with different noise levels and outlier ratios. Then, we conduct experiments using a real world data set of 2091 genes and 101 drugs in 392 cancer cell lines and compare the real experimental results from the aspect of biological process term enrichment, gene-drug and drug-drug interactions. Finally, we present interesting findings from our evaluation study and discuss the advantages and drawbacks of each method. Supplementary information: Supplementary file is available at Briefings in Bioinformatics online.

Genomic data alignment, a fundamental operation in sequencing, can be utilized to map reads into a reference sequence, query on a genomic database and perform genetic tests. However, with the reduction of sequencing cost and the accumulation of genome data, privacy-preserving genomic sequencing data alignment is becoming unprecedentedly important. In this paper, we present a comprehensive review of secure genomic data comparison schemes. We discuss the privacy threats, including adversaries and privacy attacks. The attacks can be categorized into inference, membership, identity tracing and completion attacks and have been applied to obtaining the genomic privacy information. We classify the state-of-the-art genomic privacy-preserving alignment methods into three different scenarios: large-scale reads mapping, encrypted genomic datasets querying and genetic testing to ease privacy threats. A comprehensive analysis of these approaches has been carried out to evaluate the computation and communication complexity as well as the privacy requirements. The survey provides the researchers with the current trends and the insights on the significance and challenges of privacy issues in genomic data alignment.

The significance of pan-cancer categories has recently been recognized as widespread in cancer research. Pan-cancer categorizes a cancer based on its molecular pathology rather than an organ. The molecular similarities among multi-omics data found in different cancer types can play several roles in both biological processes and therapeutic developments. Therefore, an integrated analysis for various genomic data is frequently used to reveal novel genetic and molecular mechanisms. However, a variety of algorithms for multi-omics clustering have been proposed in different fields. The comparison of different computational clustering methods in pan-cancer analysis performance remains unclear. To increase the utilization of current integrative methods in pan-cancer analysis, we first provide an overview of five popular computational integrative tools: similarity network fusion, integrative clustering of multiple genomic data types (iCluster), cancer integration via multi-kernel learning (CIMLR), perturbation clustering for data integration and disease subtyping (PINS) and low-rank clustering (LRACluster). Then, a priori interactions in multi-omics data were incorporated to detect prominent molecular patterns in pan-cancer data sets. Finally, we present comparative assessments of these methods, with discussion over key issues in applying these algorithms. We found that all five methods can identify distinct tumor compositions. The pan-cancer samples can be reclassified into several groups by different proportions. Interestingly, each method can classify the tumors into categories that are different from original cancer types or subtypes, especially for ovarian serous cystadenocarcinoma (OV) and breast invasive carcinoma (BRCA) tumors. In addition, all clusters of the five computational methods show notable prognostic values. Furthermore, both the 9 recurrent differential genes and the 15 common pathway characteristics were identified across all the methods. The results and discussion can help the community select appropriate integrative tools according to different research tasks or aims in pan-cancer analysis.

BACKGROUND: With the development of high-throughput sequencing technology, a huge amount of multi-omics data has been accumulated. Although there are many software tools for statistical analysis and visual development of omics data, these tools are not suitable for private data and non-technical users. Besides, most of these tools have specialized in only one or perhaps a few data typesare, without combining clinical information. What's more, users could not choose data processing and model selection flexibly when using these tools. RESULTS: To help non-technical users to understand and analyze private multi-omics data and ensure data security, we developed an interactive desk tool for statistical analysis and visualization of omics and clinical data (shortly IOAT). Our mainly targets csv format data, and combines clinical data with high-dimensional multi-omics data. It also contains various operations, such as data preprocessing, feature selection, risk assessment, clustering, and survival analysis. By using this tool, users can safely and conveniently try a combination of various methods on their private multi-omics data to find a model suitable for their data, conduct risk assessment and determine their cancer subtypes. At the same time, the tool can also provide them with references to genes that are closely related to tumor staging, facilitating the development of precision oncology. We review IOAT's main features and demonstrate its analysis capabilities on a lung from TCGA. CONCLUSIONS: IOAT is a local desktop tool, which provides a set of multi-omics data integration solutions. It can quickly perform a complete analysis of cancer genome data for subtype discovery and biomarker identification without security issues and writing any code. Thus, our tool can enable cancer biologists and biomedicine researchers to analyze their data more easily and safely. IOAT can be downloaded for free from https://github.com/WlSunshine/IOAT-software .

Multi-view clustering aims to partition objects into potential categories by utilizing cross-view information. One of the core issues is to sufficiently leverage different views to learn a latent subspace, within which the clustering task is performed. Recently, it has been shown that representing the multi-view data by a tensor and then learning a latent self-expressive tensor is effective. However, early works mainly focus on learning essential tensor representation from multi-view data and the resulted affinity matrix is considered as a byproduct or is computed by a simple average in Euclidean space, thereby destroying the intrinsic clustering structure. To that end, here we proposed a novel multi-view clustering method to directly learn a well-structured affinity matrix driven by the clustering task on Grassmann manifold. Specifically, we firstly employed a tensor learning model to unify multiple feature spaces into a latent low-rank tensor space. Then each individual view was merged on Grassmann manifold to obtain both an integrative subspace and a consensus affinity matrix, driven by clustering task. The two parts are modeled by a unified objective function and optimized jointly to mine a decomposable affinity matrix. Extensive experiments on eight real-world datasets show that our method achieves superior performances over other popular methods.

Converging evidence shows that disease-relevant brain alterations do not appear in random brain locations, instead, their spatial patterns follow large-scale brain networks. In this context, a powerful network analysis approach with a mathematical foundation is indispensable to understand the mechanisms of neuropathological events as they spread through the brain. Indeed, the topology of each brain network is governed by its native harmonic waves, which are a set of orthogonal bases derived from the Eigen-system of the underlying Laplacian matrix. To that end, we propose a novel connectome harmonic analysis framework that provides enhanced mathematical insights by detecting frequency-based alterations relevant to brain disorders. The backbone of our framework is a novel manifold algebra appropriate for inference across harmonic waves. This algebra overcomes the limitations of using classic Euclidean operations on irregular data structures. The individual harmonic differences are measured by a set of common harmonic waves learned from a population of individual Eigen-systems, where each native Eigen-system is regarded as a sample drawn from the Stiefel manifold. Specifically, a manifold optimization scheme is tailored to find the common harmonic waves, which reside at the center of the Stiefel manifold. To that end, the common harmonic waves constitute a new set of neurobiological bases to understand disease progression. Each harmonic wave exhibits a unique propagation pattern of neuropathological burden spreading across brain networks. The statistical power of our novel connectome harmonic analysis approach is evaluated by identifying frequency-based alterations relevant to Alzheimer's disease, where our learning-based manifold approach discovers more significant and reproducible network dysfunction patterns than Euclidean methods.

The human retinal surface resembles to a sphere while it is captured by two-dimensional (2D) planar imaging to have a stereo sequence in clinical practice. Reconstructing its three-dimensional (3D) structure from the 2D planar retinal images is crucial for analyzing the relationship between the topological morphology and clinical implication. In this regard, we propose to reconstruct the 3D retina structure from 2D stereo fundus images via dynamic registration. The fundus images from different viewpoints are first co-registrated by using multi-scale deep convolutional feature and geometric structure feature by building their transformation function. The aligned images are then mosaicked together and a 3D reconstruction is obtained by a learned weighted smoothing project the registered images onto 3D coordinates. We compare the proposed registration method with five state-of-the-art methods. Extensive experimental results demonstrate that the proposed framework achieves superior performances, even with challenging scenarios in which the tested images are severely degraded by illness, large eyeball rotation and low resolutions.

COVID-19 causes burdens to the ICU. Evidence-based planning and optimal allocation of the scarce ICU resources is urgently needed but remains unaddressed. This study aims to identify variables and test the accuracy to predict the need for ICU admission, death despite ICU care, and among survivors, length of ICU stay, before patients were admitted to ICU. Retrospective data from 733 in-patients confirmed with COVD-19 in Wuhan, China, as of March 18, 2020. Demographic, clinical and laboratory were collected and analyzed using machine learning to build the predictive models. The built machine learning model can accurately assess ICU admission, length of ICU stay, and mortality in COVID-19 patients toward optimal allocation of ICU resources. The prediction can be done by using the clinical data collected within 1-15 days before the actual ICU admission. Lymphocyte absolute value involved in all prediction tasks with a higher AUC. The online predictive system is freely available to the public (http://212.64.70.65:8000/).

Accurate nasopharyngeal carcinoma (NPC) segmentation in magnetic resonance image (MRI) is crucial for diagnosis and treatment. However, most existing deep learning methods performed unsatisfactorily, since NPC is infiltrative and typically has a small or even tiny volume with indistinguishable boundary, making it indiscernible from tightly connected surrounding tissue in immense and complex background. To address the background dominant problem, this paper proposes a coarse-to-fine deep model. The proposed model starts with predicting a coarse mask with a well-designed segmentation module, followed by a boundary rendering module, which exploits semantic information from different layers of feature maps to refine the boundary of the coarse mask. The designed rendering module is shown to achieve superior performance with dramatically fewer parameters by operating only on the segmented mask, rather than on the whole feature maps as the popular methods do. Extensive experiments are conducted on a collected dataset consisting of 2000 MRI slices from 596 patients. Experimental results demonstrate that the proposed model not only outperforms six popular segmentation models but also has a considerable generalization capability on existing models.

Multi-view clustering aims to partition the data into their underlying clusters via leveraging multiple views information. To exploit cross-view information, existed approaches in tensor-based subspace learning attract much attention. In order to explore essential tensor, the most recent work mainly focuses on capturing representation tensor with sparse and low-rank constraints. However, one shortcoming is that this process may suffer from instability since it did not consider retaining local structure between samples. To tackle the issue, we introduce a novel self-expressive tensor learning method considering both global and local constraints to promote the learning of representation tensor. In particular, we construct a tensor-based subspace representation that joint low-rank and graph-regularized tensor learning to a united optimization problem. The essential global structure and high-order correlations can be naturally captured through low-rank self-expressive tensor learning. Meanwhile, the local structures can be preserved by introducing graph regularized terms on representation tensor, thus bring benefits to subsequent clustering task. An effective optimization procedure for solving the proposed model is presented. We conduct extensive experiments on text, object, and gene expression datasets. The experimental results well demonstrate that the proposed method, named by TLGRL, achieves superiority over benchmark methods.

Identifying gene-drug patterns is a critical step in pharmacology for unveiling disease mechanisms and drug discovery. The availability of high-throughput technologies accumulates massive large-scale pharmacological and genomic data, and thus provides a new substantial opportunity to deeply understand how the oncogenic genes and the therapeutic drugs relate to each other. However, most previous studies merely used the pharmacological and genomic datasets without any prior knowledge to infer the gene-drug patterns. Here, we proposed a novel network-guided sparse binary matching model (NSBM) to decode these relationships hidden in the datasets. Not only the large-scale gene-expression data and drug-response data are jointly analyzed in our method, but also the additional prior information of genes and drugs are integrated into the form of network-based regularization. The essential structure of the NSBM model is a convex quadratic minimization problem with network-based penalties. It was demonstrated to be superior when compared with two benchmark methods through extensive experiments on both synthetic and empirical data. Posterior validation, including gene-ontology and enrichment analysis, confirmed the effectiveness of NSBM in revealing gene-drug patterns on a large-scale heterogeneous data source.

MOTIVATION: The emergence of large amounts of genomic, chemical, and pharmacological data provides new opportunities and challenges. Identifying gene-drug associations is not only crucial in providing a comprehensive understanding of the molecular mechanisms of drug action, but is also important in the development of effective treatments for patients. However, accurately determining the complex associations among pharmacogenomic data remains challenging. We propose a higher order graph matching with multiple network constraints (HOGMMNC) model to accurately identify gene-drug modules. The HOGMMNC model aims to capture the inherent structural relations within data drawn from multiple sources by hypergraph matching. The proposed technique seamlessly integrates prior constraints to enhance the accuracy and reliability of the identified relations. An effective numerical solution is combined with a novel sampling strategy to solve the problem efficiently. RESULTS: The superiority and effectiveness of our proposed method are demonstrated through a comparison with four state-of-the-art techniques using synthetic and empirical data. The experiments on synthetic data show that the proposed method clearly outperforms other methods, especially in the presence of noise and irrelevant samples. The HOGMMNC model identifies eighteen gene-drug modules in the empirical data. The modules are validated to have significant associations via pathway analysis. Significance: The modules identified by HOGMMNC provide new insights into the molecular mechanisms of drug action and provide patients with more effective treatments. Our proposed method can be applied to the study of other biological correlated module identification problems (e.g. miRNA-gene, gene-methylation, and gene-disease). AVAILABILITY AND IMPLEMENTATION: A matlab package of HOGMMNC is available at https://github.com/scutbioinformatics/HOGMMNC/. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Revealing data with intrinsically diagonal block structures is particularly useful for analyzing groups of highly correlated variables. Earlier researches based on non-negative matrix factorization (NMF) have been shown to be effective in representing such data by decomposing the observed data into two factors, where one factor is considered to be the feature and the other the expansion loading from a linear algebra perspective. If the data are sampled from multiple independent subspaces, the loading factor would possess a diagonal structure under an ideal matrix decomposition. However, the standard NMF method and its variants have not been reported to exploit this type of data via direct estimation. To address this issue, a non-negative matrix factorization with multiple constraints model is proposed in this paper. The constraints include an sparsity norm on the feature matrix and a total variational norm on each column of the loading matrix. The proposed model is shown to be capable of efficiently recovering diagonal block structures hidden in observed samples. An efficient numerical algorithm using the alternating direction method of multipliers model is proposed for optimizing the new model. Compared with several benchmark models, the proposed method performs robustly and effectively for simulated and real biological data.

The revolutionary invention of single-cell sequencing technology carves out a new way to delineate intra tumor heterogeneity and traces the evolution of single cells at the molecular level. To cater for fast and convenient needs in calling copy-number variations in analyzing single-cell sequencing data, a systematical protocol and a working pipeline is reported. The proposed pipeline consists of six modules in total. Every module in the pipeline is designed to achieve unitary task, and is unattached, thus facilitating user-customized applications. Our proposed pipeline is implemented on BGI Online to provide a user-friendly graphical interface for processing single-cell sequencing data. It consists of automatic and powerful functionalities that would facilitate the management of online Web modules for a full range of analysis, such as read alignment, genome bias correction, bin segmentation, copy number variants (CNVs) calling, data clustering, and visualization. The pipeline is open for public usage and its address is http://www.bgionline.cn.

The revolutionary invention of single-cell sequencing technology carves out a new way to delineate intra tumor heterogeneity and the evolution of single cells at the molecular level. Since single-cell sequencing requires a special genome amplification step to accumulate enough samples, a large number of bias were introduced, making the calling of copy number variants rather challenging. Accurately modeling this process and effectively detecting copy number variations (CNVs) are the major roadblock for single-cell sequencing data analysis. Recent advances manifested that the underlying copy numbers are corrupted by noise, which could be approximated by negative binomial distribution. In this paper, we formulated a general mathematical model for copy number reconstruction from read depth signal, and presented its two specific variants, namely Poisson-CNV and NB-CNV to catering for various reads distribution. Efficient numerical solution based on the classical alternating direction minimization method was designed to solve the proposed models. Extensive experiments on both synthetic datasets and empirical single-cell sequencing datasets were conducted to compare the performance of the two models. The results show that the proposed model of NB-CNV achieved superior performance in calling the CNV for single-cell sequencing data.

BACKGROUND: Variations in DNA copy number have an important contribution to the development of several diseases, including autism, schizophrenia and cancer. Single-cell sequencing technology allows the dissection of genomic heterogeneity at the single-cell level, thereby providing important evolutionary information about cancer cells. In contrast to traditional bulk sequencing, single-cell sequencing requires the amplification of the whole genome of a single cell to accumulate enough samples for sequencing. However, the amplification process inevitably introduces amplification bias, resulting in an over-dispersing portion of the sequencing data. Recent study has manifested that the over-dispersed portion of the single-cell sequencing data could be well modelled by negative binomial distributions. RESULTS: We developed a read-depth based method, nbCNV to detect the copy number variants (CNVs). The nbCNV method uses two constraints-sparsity and smoothness to fit the CNV patterns under the assumption that the read signals are negatively binomially distributed. The problem of CNV detection was formulated as a quadratic optimization problem, and was solved by an efficient numerical solution based on the classical alternating direction minimization method. CONCLUSIONS: Extensive experiments to compare nbCNV with existing benchmark models were conducted on both simulated data and empirical single-cell sequencing data. The results of those experiments demonstrate that nbCNV achieves superior performance and high robustness for the detection of CNVs in single-cell sequencing data.

Single cell sequencing provides revolutionary insight into genomic heterogeneity at the molecular level. A key procedure for successful sequencing is whole genome amplification (WGA) that permits the genomic DNA of a single cell to be amplified by several orders of magnitude. A very recently reported WGA method, Multiple Annealing and Loop-Based Amplification Cycles (MALBAC), distinguished itself by providing uniformity of amplification across the genome with low amplification bias. However, MALBAC still has limitations. For example, the polymerase enzyme is prone to introduce errors and a large proportion of single-nucleotide variations are lost. The complex biological reactions involved in MALBAC are largely unaddressed and no systematic analysis for it has been reported. To reduce both the labor and costs involved in MALBAC, we have developed a computer simulation for MALBAC. By capturing several key variables, MALBAC was formulated by a mathematical model and simulated. We compared the simulated data with empirical sequencing data of Escherichia coli genomes. The experimental results demonstrated that the coverage distribution matches very well. We believe that the proposed simulator can provide both theoretical guidance and experimental cross-validation for MALBAC, thus aiding improvement of its performance.

BACKGROUND: Classifying cancers by gene selection is among the most important and challenging procedures in biomedicine. A major challenge is to design an effective method that eliminates irrelevant, redundant, or noisy genes from the classification, while retaining all of the highly discriminative genes. RESULTS: We propose a gene selection method, called local hyperplane-based discriminant analysis (LHDA). LHDA adopts two central ideas. First, it uses a local approximation rather than global measurement; second, it embeds a recently reported classification model, K-Local Hyperplane Distance Nearest Neighbor(HKNN) classifier, into its discriminator. Through classification accuracy-based iterations, LHDA obtains the feature weight vector and finally extracts the optimal feature subset. The performance of the proposed method is evaluated in extensive experiments on synthetic and real microarray benchmark datasets. Eight classical feature selection methods, four classification models and two popular embedded learning schemes, including k-nearest neighbor (KNN), hyperplane k-nearest neighbor (HKNN), Support Vector Machine (SVM) and Random Forest are employed for comparisons. CONCLUSION: The proposed method yielded comparable to or superior performances to seven state-of-the-art models. The nice performance demonstrate the superiority of combining feature weighting with model learning into an unified framework to achieve the two tasks simultaneously.

BACKGROUND: Modeling high-dimensional data involving thousands of variables is particularly important for gene expression profiling experiments, nevertheless,it remains a challenging task. One of the challenges is to implement an effective method for selecting a small set of relevant genes, buried in high-dimensional irrelevant noises. RELIEF is a popular and widely used approach for feature selection owing to its low computational cost and high accuracy. However, RELIEF based methods suffer from instability, especially in the presence of noisy and/or high-dimensional outliers. RESULTS: We propose an innovative feature weighting algorithm, called LHR, to select informative genes from highly noisy data. LHR is based on RELIEF for feature weighting using classical margin maximization. The key idea of LHR is to estimate the feature weights through local approximation rather than global measurement, which is typically used in existing methods. The weights obtained by our method are very robust in terms of degradation of noisy features, even those with vast dimensions. To demonstrate the performance of our method, extensive experiments involving classification tests have been carried out on both synthetic and real microarray benchmark datasets by combining the proposed technique with standard classifiers, including the support vector machine (SVM), k-nearest neighbor (KNN), hyperplane k-nearest neighbor (HKNN), linear discriminant analysis (LDA) and naive Bayes (NB). CONCLUSION: Experiments on both synthetic and real-world datasets demonstrate the superior performance of the proposed feature selection method combined with supervised learning in three aspects: 1) high classification accuracy, 2) excellent robustness to noise and 3) good stability using to various classification algorithms.

We present a new method in image segmentation that is based on Otsu's method but iteratively searches for subregions of the image for segmentation, instead of treating the full image as a whole region for processing. The iterative method starts with Otsu's threshold and computes the mean values of the two classes as separated by the threshold. Based on the Otsu's threshold and the two mean values, the method separates the image into three classes instead of two as the standard Otsu's method does. The first two classes are determined as the foreground and background and they will not be processed further. The third class is denoted as a to-be-determined (TBD) region that is processed at next iteration. At the succeeding iteration, Otsu's method is applied on the TBD region to calculate a new threshold and two class means and the TBD region is again separated into three classes, namely, foreground, background, and a new TBD region, which by definition is smaller than the previous TBD regions. Then, the new TBD region is processed in the similar manner. The process stops when the Otsu's thresholds calculated between two iterations is less than a preset threshold. Then, all the intermediate foreground and background regions are, respectively, combined to create the final segmentation result. Tests on synthetic and real images showed that the new iterative method can achieve better performance than the standard Otsu's method in many challenging cases, such as identifying weak objects and revealing fine structures of complex objects while the added computational cost is minimal.

Users submit queries to an online database via its query interface. Query interface parsing, which is important for many applications, understands the query capabilities of a query interface. Since most query interfaces are organized hierarchically, we present a novel query interface parsing method, StatParser (Statistical Parser), to automatically extract the hierarchical query capabilities of query interfaces. StatParser automatically learns from a set of parsed query interfaces and parses new query interfaces. StatParser starts from a small grammar and enhances the grammar with a set of probabilities learned from parsed query interfaces under the maximum-entropy principle. Given a new query interface, the probability-enhanced grammar identifies the parse tree with the largest global probability to be the query capabilities of the query interface. Experimental results show that StatParser very accurately extracts the query capabilities and can effectively overcome the problems of existing query interface parsers.

Preservation of fine feature of an image is essential during the process of noise removal, especially via some types of smoothing such as using diffusion process-based methods to enhance images. In this paper, we present a new edge indicator called difference eigenvalue to measure image gradient magnitude in the diffusion process. Based on the eigenvalues of the Hessian matrix, the difference eigenvalue manifest itself in terms of structural information of an image. We adapt the new edge indicator to a diffusion model to achieve a better balance between noise removal and detail preservation. Experiments on both synthetic and real images show that the new model can obtain good results and outperforms existing methods.

We propose a nonlinear diffusion method based on the gradient vector field construction to remove noises in image while preserving fine details. The blocky effect and over-smoothing, as usually seen in images processed by diffusion operators, are greatly reduced by our method. Results obtained from various images, including synthetic and magnetic resonance imaging (MRI), are used to demonstrate the performance of our new method. Comparing it with other diffusion methods, we find it obtains better performance in terms of removing noises without destroying detail features of images.