Functional brain connectivity networks capture complex relationships and temporal evolution between brain regions, which have become increasingly important for diagnosing neurological disorders. However, existing methods, which are primarily based on vector or graph representations, struggle to adequately characterize the intricate spatio-temporal topological architecture of functional brain networks. Additionally, they predominantly rely on data-driven paradigms and lack priors pertaining to cross-windows network interactions. To address these issues, we propose a spatio-temporal hypergraph attention network framework for brain network analysis. Specifically, we first propose a temporal attention network architecture embedded with temporal similarity-driven prior knowledge, which effectively extracts long-range dependency information from fMRI by combining multi-head self-attention mechanisms and cross-window temporal prior knowledge. Second, we design a hierarchical hypergraph generation module that fuses local and global brain topological information to achieve multi-scale modeling of high-order spatio-temporal structures. Additionally, the spatial attention network, developed based on transformer architecture, employs hypergraph message passing mechanisms to effectively construct multi-level spatial interaction relationships between brain regions. Finally, a multi-layer perceptron (MLP) is adopted for classification. Experiments on the ADNI and PD datasets demonstrate that our method outperforms several state-of-the-art approaches in diagnostic performance and provides discriminative graph features for relevant brain disease diagnosis.

Segmenting the brain magnetic resonance (MR) images to region-of-interest (ROI) is a fundamental step for many medical image analysis tasks. Convolutional neural networks (CNNs) excel in learning the high-level contextual features for image segmentation. However, such high-level features are low-order features, which cannot reflect the complex appearance patterns of brain MR images. Intuitively, using the high-order features can enhance the performance of CNNs. Therefore, in this paper, we propose a novel Efficient Covariance Network (EfficientCovNet) that models pairwise voxel dependency features and applies it to the brain ROI segmentation tasks. Our EfficientCovNet consists of two pathways: a pairwise voxel dependency feature learning pathway that uses a novel covariance convolution to efficiently capture the pairwise features from MR images, and a contextual feature learning pathway that extracts high-level contextual features using convolutional operations. The pairwise features and contextual features are then fused together to boost brain ROI segmentation performance. Experimental results on five datasets, i.e., IXI, LONI-LPBA40, OASIS, ADNI, and CC359 datasets, demonstrate that our EfficientCovNet achieves superior performance for brain ROI segmentation in comparison with the state-of-the-art methods.

Multi-modal learning is extensively applied to diagnose brain diseases such as epilepsy and Alzheimer's disease. However, incomplete multi-modal data, where some modalities are unavailable or difficult to collect, limits the effectiveness of conventional methods. Additionally, existing approaches often overlook semantic relationships between neighbors with the same-label and latent information in missing modalities. To address these challenges, we propose an adjacent-aware distillation recovery framework designed for incomplete multi-modal learning, with a focus on diagnosing representative brain diseases, i.e. epilepsy and Alzheimer's disease. The key novelty of our framework lies in its joint design of adjacent-aware modality recovery and multi-modal representation learning in a single end-to-end pipeline. Specifically, we introduce a label-guided adjacent-aware recovery module that uses a self-attention mechanism to exploit neighbor semantics and generate distribution-consistent features for high-quality modality reconstruction. The recovered features are then refined through a knowledge distillation pathway into a modality generator, enhancing generalization under severe data incompleteness. For multi-modal representation learning, the recovered modality information is fused with the original incomplete information to enhance feature extraction and representation. Extensive experiments demonstrate the effectiveness of our method in diagnosing epilepsy and Alzheimer's disease.

Multiplex immunohistochemistry (mIHC) images have the potential to assess the complex tumor microenvironment by simultaneously detecting multiple markers within a single tissue section, however, the acquisition of mIHC images in clinical labs is both time-consuming and costly. Hence, applying machine learning-based virtual staining techniques for rapid generation of different mIHC markers has become a considerable alternative. The existing bio-image based virtual staining models generate the distributions of different markers independently, which have limited interpretability and overlook the fact that the exploration of potential interrelationships among these markers can help determine the localization of each individual marker. To address the above issues, we propose an explainable prototypical multi-task generation framework (i.e., ProtoMTG) to simultaneously generate multiple mIHC markers. Specifically, ProtoMTG involves a multi-task prototype layer that can capture the relationship among different virtual staining tasks by learning the shared and task-specific prototypes. Then, in the proto-attention layer, both task-specific and shared prototypes will be re-weighted and combined to instruct the generation of different mIHC markers. In ProtoMTG, we also design the novel prototypical activation and diversity losses to learn better prototype representation for the virtual staining task. To evaluate the performance of our method, we develop three benchmark mIHC datasets on different organs (i.e., colon, liver and stomach). The experimental results indicate that our method can not only outperform the existing image generation models, but also have good explainable ability for the virtual staining of mIHC markers. The code and dataset are available at: https://jj-zhou-code.github.io/ProtoMTG-website/.

Owing to its remarkable convenience, weak invasiveness, and strong private security, palmprint recognition has become one of the most promising biometric methods and has attracted increasing attention in both academia and industry. Although considerable recognition performance has been achieved by existing palmprint learning methods, they generally require the use of substantial labeled datasets and involve substantial computational overhead for feature learning. In this article, we propose a novel multilayer projection learning (MLPL) method to achieve efficient palmprint feature learning and recognition. First, we transform the palmprint images into their direction-specific representations by computing the difference in the multiple directional responses. Then, we learn three layers of feature projections for robust feature learning, including low-rank projection for image noise decoupling, feature projection for discriminative feature exploration, and quantization projection for information preservation during feature encoding. With multilayer feature projections, palmprint images can be transformed into discriminative feature representations through a single-step process for efficient palmprint recognition. Moreover, we extend the proposed MLPL, referred to as E-MLPL, by minimizing the representation discrepancy between heterogeneous palmprint images to make it applicable for heterogeneous palmprint recognition. The results obtained from five widely adopted databases confirm the superior performance of the proposed method in terms of both accuracy and efficiency.

The prediction of nanoparticles (NPs) distribution is crucial for the diagnosis and treatment of tumors. Recent studies indicate that the heterogeneity of tumor microenvironment (TME) highly affects the distribution of NPs across tumors. Hence, it has become a research hotspot to generate the NPs distribution by the aid of multi-modal TME components. However, the distribution divergence among multi-modal TME components may cause side effects i.e., the best unimodal model may outperform the joint generative model. To address the above issues, we propose a Divergence-Aware Multi-Modal Diffusion model (i.e., DAMM-Diffusion) to adaptively generate the prediction results from uni-modal and multi-modal branches in a unified network. In detail, the uni-modal branch is composed of the U-Net architecture while the multi-modal branch extends it by introducing two novel fusion modules i.e., Multi-Modal Fusion Module (MMFM) and Uncertainty-Aware Fusion Module (UAFM). Specifically, the MMFM is proposed to fuse features from multiple modalities, while the UAFM module is introduced to learn the uncertainty map for cross-attention computation. Following the individual prediction results from each branch, the Divergence-Aware Multi-Modal Predictor (DAMMP) module is proposed to assess the consistency of multi-modal data with the uncertainty map, which determines whether the final prediction results come from multi-modal or uni-modal predictions. We predict the NPs distribution given the TME components of tumor vessels and cell nuclei, and the experimental results show that DAMM-Diffusion can generate the distribution of NPs with higher accuracy than the comparing methods. Additional results on the multi-modal brain image synthesis task further validate the effectiveness of the proposed method. The code is released†.

Multimodal fusion provides cross-modality information to understand the human brain from different perspectives that may be missed in single modality analysis. Supervised fusion focuses on extracting multimodal patterns related to specific clinical measures by further incorporating a prior interested reference. However, existing supervised fusion methods cannot extract component that have weak correlations with the reference, which may be lost during the optimization process. Here, we propose a reference-guided parallel independent component analysis (RG-PICA) aiming at identifying multimodal covarying features related to interested reference through global optimization. The intra-modality independence, the inter-modality correlation, and the correlation between modalities and the reference are maximized globally. Simulations show that RG-PICA can accurately extract multimodal features correlated with the weak related reference while keeping cross-modality linkage comparing with seven fusion methods. In real data application, RG-PICA reveals co-varying patterns in schizophrenia (SZ) that links with cognition and correlates between modalities. These results demonstrate RG-PICA can jointly optimize for target components that correlate with the reference while keeping cross-modality linkage. This approach can improve the meaningful detection of reliable reference-linked multimodal brain patterns for brain disorders.

Adversarial attacks on palmprint recognition are crucial because these attacks can manipulate palmprint images to bypass authentication systems, posing security threats. However, many existing adversarial attacks overlook the unique features of palmprint. In this paper, we propose an enhanced palmprint adversarial attack against visible and invisible features. First, we focus on extracting palmprint main lines that are crucial for targeted adversarial attacks. Second, we introduce a channel attention mechanism that can effectively emphasize the invisible features in the palmprint image. We fuse these two features to achieve a more effective attack, ensuring that both visible and invisible details contribute to the enhancement of the adversarial attack. Finally, adversarial examples generated by our method are incorporated into the training process. The experimental results demonstrate the effectiveness of our enhanced attack.

Contactless palmprint recognition has recently emerged as a promising biometric technology. However, traditional methods that require sharing user data introduce substantial security risks. While federated learning offers privacy-preserving solutions, it often compromises recognition accuracy due to feature distribution drift caused by external factors such as lighting and devices. To address this issue, we propose an adaptive personalized federated learning framework (AdaptPFL). The central innovation lies in decomposing palmprint features into identity-related and contextual-related components using a feature decoupling mechanism. This design isolates the influence of external environmental factors on identity recognition through de-entanglement. Furthermore, two adaptive aggregation strategies are introduced to correct client drift: (1) Intra-Local Adaptive Aggregation (ILAA), which addresses intra-client drift by adaptively combining the two decoupled feature types; (2) Global-Local Adaptive Aggregation (GLAA), which corrects inter-client drift by adaptively aggregating model parameters. Experimental results demonstrate that AdaptPFL achieves superior performance compared to existing state-of-the-art methods.

Dynamic functional brain networks (DFBNs) are powerful tools in neuroscience research. Recent studies reveal that DFBNs contain heterogeneous neural nodes with more extensive connections and more drastic temporal changes, which play pivotal roles in coordinating the reorganization of the brain. Moreover, the spatio-temporal patterns of these nodes are modulated by the brain's historical states. However, existing methods not only ignore the spatio-temporal heterogeneity of neural nodes, but also fail to effectively encode the temporal propagation mechanism of heterogeneous activities. These limitations hinder the deep exploration of spatio-temporal relationships within DFBNs, preventing the capture of abnormal neural heterogeneity caused by brain diseases. To address these challenges, this paper propose a neuro-heterogeneity guided temporal graph learning strategy (NeuroH-TGL). Specifically, we first develop a spatio-temporal pattern decoupling module to disentangle DFBNs into topological consistency networks and temporal trend networks that align with the brain's operational mechanisms. Then, we introduce a heterogeneity mining module to identify pivotal heterogeneity nodes that drive brain reorganization from the two decoupled networks. Finally, we design temporal propagation graph convolution to simulate the influence of the historical states of heterogeneity nodes on the current topology, thereby flexibly extracting heterogeneous spatio-temporal information from the brain. Experiments show that our method surpasses several state-of-the-art methods, and can identify abnormal heterogeneous nodes caused by brain diseases.

Prompt learning has emerged as a promising paradigm for adapting pre-trained vision-language models (VLMs) to few-shot whole slide image (WSI) classification by aligning visual features with textual representations, thereby reducing annotation cost and enhancing model generalization. Nevertheless, existing methods typically rely on slide-level prompts and fail to capture the subtype-specific phenotypic variations of histological entities (e.g., nuclei, glands) that are critical for cancer diagnosis. To address this gap, we propose Multi-scale Attribute-enhanced Prompt Learning (MAPLE), a hierarchical framework for few-shot WSI classification that jointly integrates multi-scale visual semantics and performs prediction at both the entity and slide levels. Specifically, we first leverage large language models (LLMs) to generate entity-level prompts that can help identify multi-scale histological entities and their phenotypic attributes, as well as slide-level prompts to capture global visual descriptions. Then, an entity-guided cross-attention module is proposed to generate entity-level features, followed by aligning with their corresponding subtype-specific attributes for fine-grained entity-level prediction. To enrich entity representations, we further develop a cross-scale entity graph learning module that can update these representations by capturing their semantic correlations within and across scales. The refined representations are then aggregated into a slide-level representation and aligned with the corresponding prompts for slide-level prediction. Finally, we combine both entity-level and slide-level outputs to produce the final prediction results. Results on three cancer cohorts confirm the effectiveness of our approach in addressing few-shot pathology diagnosis tasks.

Graph Contrastive Learning (GCL) is a powerful self-supervised technique for learning node and graph representations. However, real-world graph data often exhibit imbalanced class distributions, which pose significant challenges to GCL’s effectiveness. Our experiments show that current state-of-the-art (SOTA) methods perform poorly under imbalanced settings. To address this, we propose a novel GCL framework called AIGCL for imbalanced node classification. This framework automatically and adaptively balances the node representations learned by GCL. Specifically, we introduce a new data augmentation method that retains more information from minority class nodes during graph augmentation. Additionally, we use an imbalance rate adaptive sampling strategy to balance the data. We also incorporate a Variational Graph Autoencoder (VGAE) with an encoder–decoder structure to pretrain the data and generate high-quality pseudo-labels. Our experiments demonstrate that under imbalanced settings, our model improves classification accuracy by 4 %-12 % compared to baseline models, significantly enhancing the performance of minority class nodes.

Multi-modal emotion recognition has attracted much attention in human-computer interaction, because it provides complementary information for the recognition model. However, the distribution drift among subjects and the heterogeneity of different modalities pose challenges to multi-modal emotion recognition, thereby limiting its practical application. Most of the current multi-modal emotion recognition methods are difficult to suppress above uncertainties in fusion. In this paper, we propose a cross-subject multi-modal emotion recognition framework, which jointly learns subject-independent representation and common feature between EEG and eye movements. First, we design the dynamic adversarial domain adaptation for cross-subject distribution alignment, dynamically selecting source domains in training. Second, we simultaneously capture intra-modal and inter-modal emotion-related features by both self-attention and cross-attention mechanisms, thus obtaining the robust and complementary representation of emotional information. Then, two contrastive loss functions are imposed on above network to further reduce inter-modal heterogeneity, and mine higher-order semantic similarity between synchronously collected multi-modal data. Finally, we used the output of the softmax layer as the predicted value. The experimental results on several multi-modal emotion datasets with EEG and eye movements demonstrate that our method is significantly superior to the state-of-the-art emotion recognition approaches.

Electroencephalography (EEG) biometrics draws increasing attention in high-security requirements due to its advantages of anti-spoofing, live traits, and non-duplicated. However, existing EEG datasets, which rely on external stimuli or task-specific instructions for data collection, often intertwine identity-related information with biases such as emotional states, cognitive tasks, and disease markers. Besides, EEG signals are time-varying, while identity information within EEG signals is relatively fixed, which poses challenges for extracting identity features from EEG to perform accurate person identification. This high correlation hampers the promotion of brainprint recognition in real-life applications. In this paper, we propose a disentangled representation learning based identity recognition framework, which disentangles the EEG signal into intrinsic identity-related information and biased identity-invariant information, thus enhancing the performance of EEG biometrics. First, two parallel encoders are used to extract intrinsic identity-relevant and bias identity-irrelevant factors, respectively, and each encoder consists of a temporal filter module and a novel spatial-temporal attention module. Then, we further refine the disentanglement process through a correlation-driven loss that minimizes factor similarity across spatial-temporal and global representational domains. Adversarial training and reconstruction regularization are introduced to facilitate the identity and biased representations to be independent and complementary to each other. Additionally, we extend supervised contrastive learning to the component level, minimizing cross-component similarity and encouraging each component to independently reflect its unique information, thereby improving the disentanglement efficacy. Our proposed framework achieves state-of-the-art performance on diverse datasets encompassing emotional, motor imagery, and pathological conditions, demonstrating the robustness and effectiveness of our proposed brainprint identity recognition model.

Multi-view clustering (MVC) aims to exploit the latent relationships between heterogeneous samples in an unsupervised manner, which has served as a fundamental task in the unsupervised learning community and has drawn widespread attention. In this work, we propose a new deep multi-view contrastive clustering method via graph structure awareness (DMvCGSA) by conducting both instance-level and cluster-level contrastive learning to exploit the collaborative representations of multi-view samples. Unlike most existing deep multi-view clustering methods, which usually extract only the attribute features for multi-view representation, we first exploit the view-specific features while preserving the latent structural information between multi-view data via a GCN-embedded autoencoder, and further develop a similarity-guided instance-level contrastive learning scheme to make the view-specific features discriminative. Moreover, unlike existing methods that separately explore common information, which may not contribute to the clustering task, we employ cluster-level contrastive learning to explore the clustering-beneficial consistency information directly, resulting in improved and reliable performance for the final multi-view clustering task. Extensive experimental results on twelve benchmark datasets clearly demonstrate the encouraging effectiveness of the proposed method compared with the state-of-the-art models.

Dynamic functional brain network (DFBN) can flexibly describe the time-varying topological connectivity patterns of the brain, and show great potential in brain disease diagnosis. However, most of the existing DFBN analysis methods focus on capturing the dynamic interaction at the brain region level, ignoring the spatio-temporal topological evolution across time windows. Moreover, they are difficult to suppress interfering connections in DFBNs, which leads to a diminished capacity for discerning the intrinsic structures that are intimately linked to brain disorders. To address these issues, we propose a topological evolution graph learning model to capture disease-related spatio-temporal topological features in DFBNs. Specifically, we first take the hubness of adjacent DFBN as the source domain and the target domain in turn, and then use Wasserstein distance (WD) and Gromov-Wasserstein distance (GWD) to capture the brain's evolution law at the node and edge levels, respectively. Furthermore, we introduce the principle of relevant information to guide the topology evolution graph to learn the structures that are most relevant to brain diseases yet least redundant information between adjacent DFBNs. On this basis, we develop a high-order spatio-temporal model with multi-hop graph convolution to collaboratively extract long-range spatial and temporal dependencies from the topological evolution graph. Extensive experiments show that the proposed method outperforms the current state-of-the-art methods, and can effectively reveal the information evolution mechanism between brain regions across windows.

Accurate spine segmentation is crucial in clinical diagnosis and treatment of spine diseases. However, due to the complexity of spine anatomical structure, it has remained a challenging task to accurately segment spine images. Recently, the segment anything model (SAM) has achieved superior performance for image segmentation. However, generating high-quality points and boxes is still laborious for high-dimensional medical images. Meanwhile, an accurate mask is difficult to obtain. To address these issues, in this paper, we propose a multi-atlas guided SAM using multiple pseudo mask prompts for spine image segmentation, called MA-SAM. Specifically, we first design a multi-atlas prompt generation sub-network to obtain the anatomical structure prompts. More specifically, we use a network to obtain coarse mask of the input image. Then atlas label maps are registered to the coarse mask. Subsequently, a SAM-based segmentation sub-network is used to segment images. Specifically, we first utilize adapters to fine-tune the image encoder. Meanwhile, we use a prompt encoder to learn the anatomical structure prior knowledge from the multi-atlas prompts. Finally, a mask decoder is used to fuse the image and prompt features to obtain the segmentation results. Moreover, to boost the segmentation performance, different scale features from the prompt encoder are concatenated to the Upsample Block in the mask decoder. We validate our MA-SAM on the two spine segmentation tasks, including spine anatomical structure segmentation with CT images and lumbosacral plexus segmentation with MR images. Experiment results suggest that our method achieves better segmentation performance than SAM with points, boxes, and mask prompts.

The dynamic functional brain network (DFBN) inherently captures topological changes in brain connectivity pattern during activity, attracting increasing attention for detecting brain disorders. However, most current DFBN analysis methods rely on data-driven modeling and ignore crucial prior knowledge of brain structure and function, resulting in weak interpretability of models. Furthermore, effectively extracting dynamic topological features from DFBN is still a challenging issue, due to its intricate spatio-temporal features coupling. In this paper, we propose an interpretable spatio-temporal tensor graph convolutional network for DFBN analysis. Firstly, by incorporating functional and structural priors into the construction of DBFN, we develop a hierarchical DBFN representation with brain region clustering that effectively captures the spatio-temporal topology among subnetworks. Secondly, we design a tensor graph convolutional network with both intra-graph propagation and inter-graph propagation to simultaneously extract the spatio-temporal features from the hierarchical DFBN. Additionally, we derive a functional subnetwork constraint to enhance the consistency within subnetworks and the differences between subnetworks, which guides the learned features to better reflect the topology prior of the brain network. Finally, self-attention is employed to fuse the learned dynamic topological features of different subnetworks for classification. Experimental results on epilepsy, ADNI and ABIDE datasets demonstrate that our method achieves competitive diagnostic performance and offers network-level interpretability for brain disease diagnosis.

Multi-task learning (MTL) has become a research hotspot for the analysis of whole-slide histopathological images (WSIs) since it can capture the shared representations of different tasks for the improvement of individual tasks. However, the shared representations learned by MTL are always dominated by the tasks appearing in the training set that is difficult to directly apply it on the unseen (new) tasks, especially when the unseen tasks are significantly different from the known tasks. To address the above issues, we develop a Task-Agnostic Feature-Learner (TAFL) for efficient adaptation to unseen clinical tasks, which can leverage useful image information from the existing tasks for new clinical trials with minimal task-specific modifications. Specifically, we firstly develop a neural architecture search (NAS) module that can design the network architectures of TAFL automatically. Then, a novel task-level meta-learning algorithm is developed to extract efficient and universal information from the known tasks for improving the prediction performance on the unseen tasks. We evaluate our method on three publicly available datasets derived from The Cancer Genome Atlas (TCGA) for various clinical prediction tasks (i.e., staging, cancer subtyping and survival prediction), and the experimental results indicate that our TAFL can effectively adapt to unseen tasks with better prediction performance.

Palmprint-based biometric recognition has gained widespread attention due to its rich features, contactless acquisition, and low invasiveness. However, most existing methods neglect image quality, making them less effective for low-quality, noisy palmprint images. In this paper, we propose a palm intrinsic features learning selective state space model (PalmMamba) for palmprint image denoising, which consists of shallow feature representation, noise-insensitive palmprint-specific feature learning, and sharp palmprint image restoration modules. First, we convert the degraded noisy palmprint image into a high-dimensional shallow feature representation through a single-layer convolution backbone. Then, we develop parallel learning branches, including a second-order attention-based selective state space model and a mixed difference convolution module, to exploit diverse palmprint-specific features with both global and local details. Finally, we map the fine-grained palmprint-intrinsic feature map into the identity-preserved sharp palmprint image via a commonly used convolution layer. Extensive experimental results on five public palmprint databases demonstrate the encouraging performance of the proposed PalmMamba in palmprint image denoising.

Drones have become an indispensable tool in our daily lives. While fixed-wing and rotary-wing drones are common types, each comes with its own set of advantages and disadvantages. Hybrid drones, however, combine the strengths of both types, enabling vertical takeoff and landing, hovering, and remote flight. Nonetheless, the aerodynamics of hybrid drones are exceedingly intricate, leading to a sluggish pace in their development. In this article, we propose a neural network controller design, employing the reinforcement learning Proximal Policy Optimization (PPO) algorithm to train the controller. Additionally, we integrate an attention mechanism into the network's input section to emphasize the speed variable, thereby enhancing the data processing and improving the model performance and efficiency. Experimental results demonstrate that our approach yields a controller with superior stability and optimal value.

Multi-modal emotion recognition has attracted increasing attention in human-computer interaction, as it extracts complementary information from physiological and behavioral features. Compared to single modal approaches, multi-modal fusion methods are more susceptible to uncertainty in emotion recognition, such as heterogeneity and inconsistent predictions across different modalities. Previous multi-modal approaches ignore systematic modeling of uncertainty in fusion and revelation of dynamic variations in emotion process. In this paper, we propose a dynamic confidence-aware fusion network for robust recognition of heterogeneous emotion features, including electroencephalogram (EEG) and facial expression. First, we develop a self-attention based multi-channel LSTM network to preliminarily align the heterogeneous emotion features. Second, we propose a confidence regression network to estimate true class probability (TCP) on each modality, which helps explore the uncertainty at modality level. Then, different modalities are weighted fused according to above two types of uncertainty. Finally, we adopt self-paced learning (SPL) mechanism to further improve the model robustness by alleviating negative effect from the hard learning samples. The experimental results on several multi-modal emotion datasets demonstrate the proposed method outperforms the state-of-the-art methods in emotion recognition performance and explicitly reveals the dynamic variation of emotion with uncertainty estimation. Our code is available at:

Machine learning techniques have become increasingly important in analyzing single-cell RNA and identifying cell types, providing valuable insights into cellular development and disease mechanisms. However, the presence of batch effects poses major challenges in scRNA-seq analysis due to data distribution variation across batches. Although several batch effect mitigation algorithms have been proposed, most of them focus only on the correlation of local structure embeddings, ignoring global distribution matching and discriminative feature representation in batch correction. In this paper, we proposed the discriminative domain adaption network (D2AN) for joint batch effects correction and type annotation with single-cell RNA-seq. Specifically, we first captured the global low-dimensional embeddings of samples from the source and target domains by adversarial domain adaption strategy. Second, a contrastive loss is developed to preliminarily align the source domain samples. Moreover, the semantic alignment of class centroids in the source and target domains is achieved for further local alignment. Finally, a self-paced learning mechanism based on inter-domain loss is adopted to gradually select samples with high similarity to the target domain for training, which is used to improve the robustness of the model. Experimental results demonstrated that the proposed method on multiple real datasets outperforms several state-of-the-art methods.

Brain region-of-interest (ROI) segmentation with magnetic resonance (MR) images is a basic prerequisite step for brain analysis. The main problem with using deep learning for brain ROI segmentation is the lack of sufficient annotated data. To address this issue, in this paper, we propose a simple multi-atlas supervised contrastive learning framework (MAS-CL) for brain ROI segmentation with MR images in an end-to-end manner. Specifically, our MAS-CL framework mainly consists of two steps, including 1) a multi-atlas supervised contrastive learning method to learn the latent representation using a limited amount of voxel-level labeling brain MR images, and 2) brain ROI segmentation based on the pre-trained backbone using our MSA-CL method. Specifically, different from traditional contrastive learning, in our proposed method, we use multi-atlas supervised information to pre-train the backbone for learning the latent representation of input MR image, i.e., the correlation of each sample pair is defined by using the label maps of input MR image and atlas images. Then, we extend the pre-trained backbone to segment brain ROI with MR images. We perform our proposed MAS-CL framework with five segmentation methods on LONI-LPBA40, IXI, OASIS, ADNI, and CC359 datasets for brain ROI segmentation with MR images. Various experimental results suggested that our proposed MAS-CL framework can significantly improve the segmentation performance on these five datasets.

Brain networks, describing the functional or structural interactions of brain with graph theory, have been widely used for brain imaging analysis. Currently, several network representation methods have been developed for describing and analyzing brain networks. However, most of these methods ignored the valuable weighted information of the edges in brain networks. In this paper, we propose a new representation method (i.e., ordinal pattern tree) for brain network analysis. Compared with the existing network representation methods, the proposed ordinal pattern tree (OPT) can not only leverage the weighted information of the edges but also express the hierarchical relationships of nodes in brain networks. On OPT, nodes are connected by ordinal edges which are constructed by using the ordinal pattern relationships of weighted edges. We represent brain networks as OPTs and further develop a new graph kernel called optimal transport (OT) based ordinal pattern tree (OT-OPT) kernel to measure the similarity between paired brain networks. In OT-OPT kernel, the OT distances are used to calculate the transport costs between the nodes on the OPTs. Based on these OT distances, we use exponential function to calculate OT-OPT kernel which is proved to be positive definite. To evaluate the effectiveness of the proposed method, we perform classification and regression experiments on ADHD-200, ABIDE and ADNI datasets. The experimental results demonstrate that our proposed method outperforms the state-of-the-art graph methods in the classification and regression tasks.

With the remarkable success of digital histopathology and the deep learning technology, many whole-slide pathological images (WSIs) based deep learning models are designed to help pathologists diagnose human cancers. Recently, rather than predicting categorical variables as in cancer diagnosis, several deep learning studies are also proposed to estimate the continuous variables such as the patients' survival or their transcriptional profile. However, most of the existing studies focus on conducting these predicting tasks separately, which overlooks the useful intrinsic correlation among them that can boost the prediction performance of each individual task. In addition, it is sill challenge to design the WSI-based deep learning models, since a WSI is with huge size but annotated with coarse label. In this study, we propose a general multi-instance multi-task learning framework (HistMIMT) for multi-purpose prediction from WSIs. Specifically, we firstly propose a novel multi-instance learning module (TMICS) considering both common and specific task information across different tasks to generate bag representation for each individual task. Then, a soft-mask based fusion module with channel attention (SFCA) is developed to leverage useful information from the related tasks to help improve the prediction performance on target task. We evaluate our method on three cancer cohorts derived from the Cancer Genome Atlas (TCGA). For each cohort, our multi-purpose prediction tasks range from cancer diagnosis, survival prediction and estimating the transcriptional profile of gene TP53. The experimental results demonstrated that HistMIMT can yield better outcome on all clinical prediction tasks than its competitors.

Dynamic brain network has the advantage over static brain network in characterizing the variation pattern of functional brain connectivity, and it has attracted increasing attention in brain disease diagnosis. However, most of the existing dynamic brain networks analysis methods rely on extracting features from independent brain networks divided by sliding windows, making them hard to reveal the high-order dynamic evolution laws of functional brain networks. Additionally, they cannot effectively extract the spatio-temporal topology features in dynamic brain networks. In this paper, we propose to use optimal transport (OT) theory to capture the topology evolution of the dynamic brain networks, and develop a multi-channel spatio-temporal graph convolutional network that collaboratively extracts the temporal and spatial features from the evolution networks. Specifically, we first adaptively evaluate the graph hubness of brain regions in the brain network of each time window, which comprehensively models information transmission among multiple brain regions. Second, the hubness propagation information across adjacent time windows is captured by optimal transport, describing high-order topology evolution of dynamic brain networks. Moreover, we develop a spatio-temporal graph convolutional network with attention mechanism to collaboratively extract the intrinsic temporal and spatial topology information from the above networks. Finally, the multi-layer perceptron is adopted for classifying the dynamic brain network. The extensive experiment on the collected epilepsy dataset and the public ADNI dataset show that our proposed method not only outperforms several state-of-the-art methods in brain disease diagnosis, but also reveals the key dynamic alterations of brain connectivities between patients and healthy controls.

Palmprint has attracted increasing attention for biometric recognition in recent years due to its outstanding reliability, user-friendliness and hygiene. However, existing palmprint recognition methods usually require high-quality palmprint images with clear texture and line patterns; however, in practical applications palmprint images are usually of low quality. In this study, we propose a dense hybrid attention (DHA) network for palmprint image super-resolution (SR) by recovering the clear palmprint-specific characteristics. The proposed DHA network first obtains the high-dimensional shallow representation via a single convolution layer, and then jointly learns the local and global palmprint-specific features via parallel convolutional neural network (CNN)-and transformer-based branches. Particularly, we develop two enhanced spatial and channel attention (CA) modules to adaptively emphasize the local position-specific characteristics of palmprints, such that the SR palmprint images can be well recovered with clear texture and edge characteristics. Experimental results on three publicly used palmprint databases clearly show the effectiveness of the proposed method for palmprint image SR.

Abstract A first‐order finite‐state Markov chain (FSMC) typically models the Rayleigh fading channel in the open literature because the first‐order FSMC is analytically tractable and can derive closed‐form results. Non‐orthogonal multiple access (NOMA) has been recognised as a novel wireless technology that addresses challenges in the next generation of mobile communications. According to the power‐domain NOMA protocol, channels in the NOMA wireless network are sorted by the channel gain. Then considering NOMA, there is insufficient information on how to further form a suitable model for ordered Rayleigh fading channels based on the first‐order FSMC. Given the mathematical statement on how to model the order statistics of multidimensional Markov chains for ordered Rayleigh fading channels, the authors consider these order statistics as a Markov chain, and propose specific processes of representing the state space and constructing the transition probability matrix accordingly. Numerical and simulation results validate the mathematical correctness and accuracy of these novel processes. In addition, for ordered Rayleigh fading channels, the performances of various methods of partitioning the entire signal‐to‐noise ratio range are compared. The performance comparison results are the same as those obtained for the individual unordered Rayleigh fading channel.

The multi-site approach has attracted increasing attention in brain disease diagnosis, because it can improve the prediction performance by integrating sample information from different medical institutions. However, its training procedure requires the transmission of subject's original images or features among sites, which may cause privacy disclosure. In this paper, we propose a self-supervised federated adaptation (S2FA) framework for robust multi-site prediction, which can reduce the risk of privacy disclosure. As far as we know, it is the first work to investigate the cross-site brain disease diagnosis, which trains model on source sites and tests on target site, often occurring in clinical practice. Firstly, we implement a decentralized federated optimization strategy, by which each site communicates model parameters periodically. Secondly, we construct an auxiliary self-supervised model for target site through transferring knowledge from source sites with self-paced learning. Then, a hash mapping is proposed to encode the target feature, simultaneously reducing the risk of privacy information disclosure and alleviating data heterogeneity among sites. Finally, we achieve the cross-site prediction by weighted federated source model and auxiliary target model. Experimental results on multi-site datasets show that the proposed S2FA can accurately identify brain disease. Our codes are available athttps://github.com/nuaayqm/S2FA.

Multi-center analysis has attracted increasing attention in brain disease diagnosis, because it provides effective approaches to improve disease diagnostic performance by making use of the information from different centers. However, in practical multi-center applications, data uncertainty is more common than that in single center, which brings challenge to robust modeling of diagnosis. In this article, we proposed a multi-discriminator active adversarial network (MDAAN) to alleviate the uncertainties at the center, feature, and label levels for multi-center brain disease diagnosis. First, we extract the latent invariant representation of the source center and target center to reduce domain shift by adversarial learning strategy. Second, the proposed method adaptively evaluates the contribution of different source centers in fusion by measuring data distribution difference between source and target center. Moreover, only the hard learning samples in target center are identified to label with low sample annotation cost. Finally, we treat the selected samples as the auxiliary domain to alleviate the negative transfer and improve the robustness of the multi-center model. We extensively compare the proposed approach with several state-of-the-art multi-center methods on the five-center schizophrenia dataset, and the results demonstrate that our method is superior to the previous methods in identifying brain disease.

Contactless palmprint recognition, an emerging biometric technology, has attracted increasing attention due to its noninvasive and high practicability characteristics. Although it is naturally suitable for mobile application scenarios, the following two challenges severely limit its recognition performance: 1) the inconsistency in acquisition devices used in training and testing, and 2) many subjects are unable to be imaged on each device, resulting in incomplete data problems. To address these issues, we propose a self-paced CycleGAN with self-attention modules, which simultaneously synthesizes missing data and alleviates the influence of different imaging devices. Specifically, we develop CycleGAN with self-attention modules to generate missing training data by effectively mining the structural correlation among samples while capturing the cross-domain features. Furthermore, a self-paced learning strategy, which is a human cognitive-driven learning mechanism, is used to guide learning the robust cross-domain feature representation and recognition model, by which the relatively easy learning samples are gradually involved in the training process. To verify the effectiveness of the proposed method, we conduct experiments on contactless palmprint datasets collected using different smartphones. The results show that our approach outperforms state-of-the-art methods in classifying contactless palmprint images.

As an emerging biometric technology, multi-spectral palmprint recognition has attracted increasing attention in security due to its high accuracy and ease of use. Compared to single spectral case, multi-spectral palmprint model is more susceptible to the attack of adversarial examples. However, the previous adversarial example attack approaches cannot generate the most aggressive adversarial examples for multi-spectral palmprint recognition. In addition, most of them are dependent on the explicit architecture or need time-consuming queries about the network to be attacked, which significantly limits their application in the field of security. To solve the above problems, in this paper, we proposed the multi-spectral palmprints joint attack and defense framework based on multi-view adversarial examples learning. First, we respectively capture the multi-view deep common feature space for the different spectra and the discriminative feature space across the different subjects. Second, we introduce perturbation in the deep common space to achieve adversarial multi-spectral palmprints with gradient propagation. In addition, we pursue the manifold of the difference space and use it to suppress the discriminability of the recognition model with adversarial region theory. Finally, the generated adversarial examples are fed into the training model to enhance the robustness of the recognition algorithm. The experimental results on multi-spectral palmprint dataset demonstrate that the proposed multi-view joint attack approach is superior to the state-of-the-art adversarial example attack methods in attack accuracy and transferability. Moreover, the defense strategy with the adversarial examples by our method can significantly promote the robustness of multi-spectral palmprint recognition methods.

Survival analysis is to estimate the survival time for an individual or a group of patients, which is a valid solution for cancer treatments. Recent studies suggested that the integrative analysis of histopathological images and genomic data can better predict the survival of cancer patients than simply using single bio-marker, for different bio-markers may provide complementary information. However, for the given multi-modal data that may contain irrelevant or redundant features, it is still challenge to design a distance metric that can simultaneously discover significant features and measure the difference of survival time among different patients. To solve this issue, we propose a Feature-Aware Multi-modal Metric Learning method (FAM3L), which not only learns the metric for distance constraints on patients' survival time, but also identifies important images and genomic features for survival analysis. Specifically, for each modality of data, we firstly design one feature-aware metric that can be decoupled into a traditional distance metric and a diagonal weight for important feature identification. Then, in order to explore the complex correlation across multiple modality data, we apply Hilbert-Schmidt Independence Criterion (HSIC) to jointly learn multiple metrics. Finally, based on the learned distance metrics, we apply the Cox proportional hazards model for prognosis prediction. We evaluate the performance of our proposed FAM3L method on three cancer cohorts derived from The Cancer Genome Atlas (TCGA), the experimental results demonstrate that our method can not only achieve superior performance for cancer prognosis, but also identify meaningful image and genomic features correlating strongly with cancer survival.

The tumor-infiltrating lymphocytes (TILs) and its correlation with tumors have shown significant values in the development of cancers. Many observations indicated that the combination of the whole-slide pathological images (WSIs) and genomic data can better characterize the immunological mechanisms of TILs. However, the existing image-genomic studies evaluated the TILs by the combination of pathological image and single-type of omics data (e.g., mRNA), which is difficulty in assessing the underlying molecular processes of TILs holistically. Additionally, it is still very challenging to characterize the intersections between TILs and tumor regions in WSIs and the high dimensional genomic data also brings difficulty for the integrative analysis with WSIs. Based on the above considerations, we proposed an end-to-end deep learning framework i.e., IMO-TILs that can integrate pathological image with multi-omics data (i.e., mRNA and miRNA) to analyze TILs and explore the survival-associated interactions between TILs and tumors. Specifically, we firstly apply the graph attention network to describe the spatial interactions between TILs and tumor regions in WSIs. As to genomic data, the Concrete AutoEncoder (i.e., CAE) is adopted to select survival-associated Eigengenes from the high-dimensional multi-omics data. Finally, the deep generalized canonical correlation analysis (DGCCA) accompanied with the attention layer is implemented to fuse the image and multi-omics data for prognosis prediction of human cancers. The experimental results on three cancer cohorts derived from the Cancer Genome Atlas (TCGA) indicated that our method can both achieve higher prognosis results and identify consistent imaging and multi-omics bio-markers correlated strongly with the prognosis of human cancers.

Multi-modal fusion has become an important data analysis technology in Alzheimer's disease (AD) diagnosis, which is committed to effectively extract and utilize complementary information among different modalities. However, most of the existing fusion methods focus on pursuing common feature representation by transformation, and ignore discriminative structural information among samples. In addition, most fusion methods use high-order feature extraction, such as deep neural network, by which it is difficult to identify biomarkers. In this paper, we propose a novel method named deep multi-modal discriminative and interpretability network (DMDIN), which aligns samples in a discriminative common space and provides a new approach to identify significant brain regions (ROIs) in AD diagnosis. Specifically, we reconstruct each modality with a hierarchical representation through multilayer perceptron (MLP), and take advantage of the shared self-expression coefficients constrained by diagonal blocks to embed the structural information of inter-class and the intra-class. Further, the generalized canonical correlation analysis (GCCA) is adopted as a correlation constraint to generate a discriminative common space, in which samples of the same category gather while samples of different categories stay away. Finally, in order to enhance the interpretability of the deep learning model, we utilize knowledge distillation to reproduce coordinated representations and capture influence of brain regions in AD classification. Experiments show that the proposed method performs better than several state-of-the-art methods in AD diagnosis.

Domain adaptation leverages rich knowledge from a related source domain so that it can be used to perform tasks in a target domain. For more knowledge to be obtained under relaxed conditions, domain adaptation methods have been widely used in pattern recognition and image classification. However, most of the existing domain adaptation methods only consider how to minimize different distributions of the source and target domains, which neglects what should be transferred for a specific task and suffers negative transfer by distribution outliers. To address these problems, in this paper, we propose a novel domain adaptation method called weighted correlation embedding learning (WCEL) for image classification. In the WCEL approach, we seamlessly integrated correlation learning, graph embedding, and sample reweighting into a unified learning model. Specifically, we extracted the maximum correlated features from the source and target domains for image classification tasks. In addition, two graphs were designed to preserve the discriminant information from interclass samples and neighborhood relations in intraclass samples. Furthermore, to prevent the negative transfer problem, we developed an efficient sample reweighting strategy to predict the target with different confidence levels. To verify the performance of the proposed method in image classification, extensive experiments were conducted with several benchmark databases, verifying the superiority of the WCEL method over other state-of-the-art domain adaptation algorithms.

Multi-modal imaging data fusion has attracted much attention in medical data analysis because it can provide complementary information for more accurate analysis. Integrating functional and structural multi-modal imaging data has been increasingly used in the diagnosis of brain diseases, such as epilepsy. Most of the existing methods focus on the feature space fusion of different modalities but ignore the valuable high-order relationships among samples and the discriminative fused features for classification. In this paper, we propose a novel framework by fusing data from two modalities of functional MRI (fMRI) and diffusion tensor imaging (DTI) for epilepsy diagnosis, which effectively captures the complementary information and discriminative features from different modalities by high-order feature extraction with the attention mechanism. Specifically, we propose a triple network to explore the discriminative information from the high-order representation feature space learned from multi-modal data. Meanwhile, self-attention is introduced to adaptively estimate the degree of importance between brain regions, and the cross-attention mechanism is utilized to extract complementary information from fMRI and DTI. Finally, we use the triple loss function to adjust the distance between samples in the common representation space. We evaluate the proposed method on the epilepsy dataset collected from Jinling Hospital, and the experiment results demonstrate that our method is significantly superior to several state-of-the-art diagnosis approaches.

In recent years, numerous studies have adopted rs-fMRI to construct dynamic functional connectivity networks (DFCNs) and applied them to the diagnosis of brain diseases, such as epilepsy and schizophrenia. Compared with the static brain networks, the DFCNs have a natural advantage in reflecting the process of brain activity due to the time information contained in it. However, most of the current methods for constructing DFCNs fail to aggregate the brain topology structure and temporal variation of the functional architecture associated with brain regions, and often ignore the inherent multi-dimensional feature representation of DFCNs for classification. In order to address these issues, we propose a novel DFCNs construction and representation method and apply it to brain disease diagnosis. Specifically, we fuse the blood oxygen level dependent (BOLD) signal and interactions between brain regions to distinguish the brain topology within each time domain and across different time domains, by embedding block structure in the adjacency matrix. After that, a sparse tensor decomposition method with sparse local structure preserving regularization is developed to extract DFCNs features from a multi-dimensional perspective. Finally, the kernel discriminant analysis is employed to provide the decision result. We validate the proposed method on epilepsy and schizophrenia identification tasks, respectively. The experimental results show that the proposed method outperforms several state-of-the-art methods in the diagnosis of brain diseases.

Currently, how to conjointly fuse structural connectivity (SC) and functional connectivity (FC) for identifying brain diseases is a hot topic in the area of brain network analysis. Most of the existing works combine two types of connectivity in decision level, thus ignoring the underlying relationship between SC and FC. To solve this problem, in this paper, we model the brain network as the multi-layer network formed by the SC and FC, and then propose a coherent pattern to represent structural information of the multi-layer network for the brain disease identification. The proposed coherent pattern consists of a paired-subgraph extracted from the FC and SC within the same node-set. Compared with the previous methods, this coherent pattern not only describes the connectivity information of both SC and FC by subgraphs at each layer, but also reflects their intrinsic relationship by the co-occurrence pattern of the paired-subgraph. Based on this coherent pattern, we further develop a framework for identifying brain diseases. Specifically, we first construct multi-layer networks by using SC and FC for each subject and then mine coherent patterns that frequently appear in each group. Next, we select the discriminative coherent pattern from these frequent coherent patterns according to their frequency of occurrence. Finally, we construct a feature matrix for each subject based on the binary indicator vector and then use the support vector machine (SVM) as its classifier. Experimental results on real epilepsy datasets demonstrate that our method outperforms several state-of-the-art approaches in the tasks of brain disease classification.

The fractional amplitude of low-frequency fluctuations (fALFF) has been widely used as potential clinical biomarkers for resting-state functional-magnetic-resonance-imaging-based schizophrenia diagnosis. How-ever, previous studies usually measure the fALFF with specific bands from 0.01 to 0.08 Hz, which cannot fully delineate the complex variations of spontaneous fluctuations in the resting-state brain. In addition, fALFF data are intrinsically constrained by the brain structure, but most of the traditional methods have not consider it in feature selection. For addressing these problems, we propose a model to classify schizophrenia in multifrequency bands with tree-guided group sparse learning. In detail, we first acquire the fALFF data in multifrequency bands (i.e., slow-5: 0.01-0.027 Hz, slow-4: 0.027-0.073 Hz, slow-3: 0.073-0.198 Hz, and slow-2: 0.198-0.25 Hz). Then, we divide the whole brain into different candidate patches and select those significant patches related to schizophrenia using random forest-based important score. Moreover, we use tree-structured sparse learning method for feature selection with the above patch spatial constraint. Finally, considering biomarkers from multifrequency bands can reflect complementary information among multiple-frequency bands, we adopt the multikernel learning method to combine features of multifrequency bands for classification. Our experimental results show that these biomarkers from multifrequency bands can achieve a classification accuracy of 91.1% on 17 schizophrenia patients and 17 healthy controls, further demonstrating that the multifrequency bands analysis can better account for classification of schizophrenia.

Sparse representation based classification has attracted much attention due to its robustness in the fields of biometrics, such as face recognition, palm-print recognition. SRC first constructs the linear representation model for the test sample, and then it classifies the test sample by comparing its coding error of each class. The coding error essentially can be viewed as the distance from sample to the subspace spanned by the samples from specific class. Therefore, the key to improve the classification performance of SRC is to enhance the subspace separability of these subspaces. In this paper, we introduce the difference subspace analysis into SRC, and propose difference subspace based SRC (DSSRC) for face recognition. Different from traditional dictionary learning based SRC methods, DSSRC focuses on maximizing the discriminability for the classes rather than the representation ability for the samples. Extensive experiments on the well-known image datasets demonstrate that the proposed DSSRC method is effective for face recognition.

In the field of face recognition, conventional dictionary learning algorithms mainly focus on reconstructing the training samples and cannot directly associate the learning procedure with the test samples. Thus, they may not well represent the test samples and obtain unsatisfactory classification p erformance. In addition, though different training samples have various contributions to learn a dictionary, conventional dictionary learning algorithms cannot well exploit these contributions. In order to address these problems, we present a test sample oriented two-phase dictionary learning (TSOTP-DL) algorithm for face recognition. In the first phase of the TSOTP-DL algorithm, we use all training samples to provide a linear representation of the test sample, and select K ``important'' training samples by using the variety of contributions. In the second phase of the TSOTP-DL algorithm, a dictionary is learned for the test sample by using the selected K$ ``important'' training samples. The TSOTP-DL algorithm utilizes the testing sample to select a subset of the training samples for learning a dictionary, which can reduce the influence of noise. Thus, the training samples are refined according to their contributions to the test sample in our algorithm, and it can improve the discriminative ability of the learned dictionary. In order to further improve the discriminative ability of the learned dictionary, a label embedding of atoms is constructed to encourage the same class training samples to have more similar coding coefficients than different classes. Experiment results demonstrate that our proposed algorithm achieves better classification results than some state-of-the-art dictionary learning and sparse coding algorithms on four public face databases.

With the development of deep sequencing technologies, many RNA-Seq data have been generated. Researchers have proposed many methods based on the sparse theory to identify the differentially expressed genes from these data. In order to improve the performance of sparse principal component analysis, in this paper, we propose a novel class-information-based sparse component analysis (CISCA) method which introduces the class information via a total scatter matrix. First, CISCA normalizes the RNA-Seq data by using a Poisson model to obtain their differential sections. Second, the total scatter matrix is gotten by combining the between-class and within-class scatter matrices. Third, we decompose the total scatter matrix by using singular value decomposition and construct a new data matrix by using singular values and left singular vectors. Then, aiming at obtaining sparse components, CISCA decomposes the constructed data matrix by solving an optimization problem with sparse constraints on loading vectors. Finally, the differentially expressed genes are identified by using the sparse loading vectors. The results on simulation and real RNA-Seq data demonstrate that our method is effective and suitable for analyzing these data.

Sparse representation for classification (SRC) has attracted much attention in recent years. It usually performs well under the following assumptions. The first assumption is that each class has sufficient training samples. In other words, SRC is not good at dealing with the undersampled problem, i.e., each class has few training samples, even single sample. The second one is that the sample vectors belonging to different classes should not distribute on the same vector direction. However, the above two assumptions are not always satisfied in real-world problems. In this paper, we propose a novel SRC based algorithm, i.e., kernel sparse representation based classifier for undersampled problem (KSRC-UP) to perform classification. It does not need the above assumptions in principle. KSRC-UP can deal well with the small scale and high dimensional real world data sets. Experiments on the popular face databases show that our KSRC-UP method can perform better than other SRC methods.