Mining Module Research Articles

PPBI: Pose-Guided Partial-Attention Network with Batch Information for Occluded Person Re-Identification

Occludedperson re-identification (ReID) tasks pose a significant challenge in matching occluded pedestrians to their holistic counterparts across diverse camera views and scenarios. Robust representational learning is crucial in this context, given the unique challenges introduced by occlusions. Firstly, occlusions often result in missing or distorted appearance information, making accurate feature extraction difficult. Secondly, most existing methods focus on learning representations from isolated images, overlooking the potential relational information within image batches. To address these challenges, we propose a pose-guided partial-attention network with batch information (PPBI), designed to enhance both spatial and relational learning for occluded ReID tasks. PPBI includes two core components: (1) A node optimization network (NON) that refines the relationships between key-point nodes of a pedestrian to better address occlusion-induced inconsistencies. (2) A key-point batch attention (KBA) module that explicitly models inter-image interactions across batches to mitigate occlusion effects. Additionally, we introduce a correction of hard mining (CHM) module to handle occlusion-related misclassification and a batch enhancement (BE) model to strengthen key-point attention across image batches. Extensive experiments on occluded and holistic ReID tasks validate the effectiveness of PPBI. Our framework achieves a 2.7% mAP improvement over HoNeT on the Occluded-Duke dataset, demonstrating its robust performance.

GLCANet: Global–Local Context Aggregation Network for Cropland Segmentation from Multi-Source Remote Sensing Images

Cropland is a fundamental basis for agricultural development and a prerequisite for ensuring food security. The segmentation and extraction of croplands using remote sensing images are important measures and prerequisites for detecting and protecting farmland. This study addresses the challenges of diverse image sources, multi-scale representations of cropland, and the confusion of features between croplands and other land types in large-area remote sensing image information extraction. To this end, a multi-source self-annotated dataset was developed using satellite images from GaoFen-2, GaoFen-7, and WorldView, which was integrated with public datasets GID and LoveDA to create the CRMS dataset. A novel semantic segmentation network, the Global–Local Context Aggregation Network (GLCANet), was proposed. This method integrates the Bilateral Feature Encoder (BFE) of CNNs and Transformers with a global–local information mining module (GLM) to enhance global context extraction and improve cropland separability. It also employs a multi-scale progressive upsampling structure (MPUS) to refine the accuracy of diverse arable land representations from multi-source imagery. To tackle the issue of inconsistent features within the cropland class, a loss function based on hard sample mining and multi-scale features was constructed. The experimental results demonstrate that GLCANet improves OA and mIoU by 3.2% and 2.6%, respectively, compared to the existing advanced networks on the CRMS dataset. Additionally, the proposed method also demonstrated high precision and practicality in segmenting large-area croplands in Chongzhou City, Sichuan Province, China.

High-Quality Instance Mining and Weight Re-Assigning for Weakly Supervised Object Detection in Remote Sensing Images

Weakly supervised object detection (WSOD) in remote sensing images (RSIs) achieves high-value object classification and localization by only using image-level labels. However, two problems limit its performance. Firstly, adjacent instances are often misclassified because their pseudo-labels are determined solely based on the spatial distances between them and their corresponding seed instances. Secondly, most WSOD methods assign the highest weight to the instance that covers the discriminative part of an object, thereby urging WSOD models to focus on the discriminative part rather than the whole object. To handle the first problem, a high-quality instance mining (HQIM) module, which incorporates the feature similarities between instances into the label propagation process, enabling some misclassified adjacent instances to be removed, is proposed. To tackle the second problem, a weight re-assigning (WRA) strategy, which redistributes the loss weights of instances, is proposed. Specifically, the loss weights of instances focusing on the discriminative part are exchanged with those of instances that broadly cover the whole object. Ablation studies demonstrate the effectiveness of HQIM and WRA, while comparisons with popular models on two RSI benchmarks further verify the effectiveness of our model.

Confusion Region Mining for Crowd Counting.

Existing works mainly focus on crowd and ignore the confusion regions which contain extremely similar appearance to crowd in the background, while crowd counting needs to face these two sides at the same time. To address this issue, we propose a novel end-to-end trainable confusion region discriminating and erasing network called CDENet. Specifically, CDENet is composed of two modules of confusion region mining module (CRM) and guided erasing module (GEM). CRM consists of basic density estimation (BDE) network, confusion region aware bridge and confusion region discriminating network. The BDE network first generates a primary density map, and then the confusion region aware bridge excavates the confusion regions by comparing the primary prediction result with the ground-truth density map. Finally, the confusion region discriminating network learns the difference of feature representations in confusion regions and crowds. Furthermore, GEM gives the refined density map by erasing the confusion regions. We evaluate the proposed method on four crowd counting benchmarks, including ShanghaiTech Part_A, ShanghaiTech Part_B, UCF_CC_50, and UCF-QNRF, and our CDENet achieves superior performance compared with the state-of-the-arts.

PIM-Net: Progressive Inconsistency Mining Network for image manipulation localization

The content authenticity and reliability of digital images have promoted the research on image manipulation localization (IML). Most current deep learning-based methods focus on extracting global or local tampering features for identifying forged regions. These features usually contain semantic information and lead to inaccurate detection results for non-object or incomplete semantic tampered regions. In this study, we propose a novel Progressive Inconsistency Mining Network (PIM-Net) for effective IML. Specifically, PIM-Net consists of two core modules, the Inconsistency Mining Module (ICMM) and the Progressive Fusion Refinement module (PFR). ICMM models the inconsistency between authentic and forged regions at two levels, i.e., pixel correlation inconsistency and region attribute incongruity, while avoiding the interference of semantic information. Then PFR progressively aggregates and refines extracted inconsistent features, which in turn yields finer and pure localization responses. Extensive qualitative and quantitative experiments on five benchmarks demonstrate PIM-Net’s superiority to current state-of-the-art IML methods. Code: https://github.com/ningnbai/PIM-Net.

One point is all you need for weakly supervised object detection

Object detection with weak annotations has attracted much attention recently. Weakly supervised object detection(WSOD) methods which only use image-level labels to train a detector encounter some severe problems that it cannot cover the whole object and the region proposal methods waste a large amount of time. Meanwhile, point supervised object detection(PSOD) leverages point annotations that remarkably improves the performance. However, point annotation is still complex and increase the costs of annotation. To overcome these issues, we propose a novel method which only requires one point per category for a training image. Compared to point annotation, our method significantly reduces the annotation cost as the number of point annotations is largely reduced. We design a framework to train a detector with one point per category annotation. Firstly, a pseudo box generation module is introduced to generate the corresponding pseudo boxes of the annotated points. Then, inspired by the observation that the features of objects with the same class in an image are very similar, a dense instances mining module is proposed to make use of the similarity between the features of objects with the same class to discover unlabeled instances and generate pseudo category heatmaps. Finally, the pseudo boxes and pseudo category heatmaps are leveraged to train a detector. Experiments conducted on popular open-source datasets verify the effectiveness of our annotation method and framework. Our proposed method outperforms previous WSOD methods and achieves comparable performance with some PSOD methods in a more efficient way.

FDT‐Net: Frequency‐Aware Dual‐Branch Transformer‐Based Optic Cup and Optic Disk Segmentation With Parallel Contour Information Mining and Uncertainty‐Guided Refinement

ABSTRACTAccurate segmentation of the optic cup and disc in fundus images is crucial for the prevention and diagnosis of glaucoma. However, challenges arise due to factors such as blood vessels, and mainstream networks often demonstrate limited capacity in extracting contour information. In this paper, we propose a segmentation framework named FDT‐Net, which is based on a frequency‐aware dual‐branch Transformer (FDBT) architecture with parallel contour information mining and uncertainty‐guided refinement. Specifically, we design a FDBT that operates in the frequency domain. This module leverages the inherent contextual awareness of Transformers and utilizes Discrete Cosine Transform (DCT) transformation to mitigate the impact of certain interference factors on segmentation. The FDBT comprises global and local branches that independently extract global and local information, thereby enhancing segmentation results. Moreover, to further mine additional contour information, this study develops the parallel contour information mining (PCIM) module to operate in parallel. These modules effectively capture more details from the edges of the optic cup and disc while avoiding mutual interference, thus optimizing segmentation performance in contour regions. Furthermore, we propose an uncertainty‐guided refinement (UGR) module, which generates and quantifies uncertainty mass and leverages it to enhance model performance based on subjective logic theory. The experimental results on two publicly available datasets demonstrate the superior performance and competitive advantages of our proposed FDT‐Net. The code for this project is available at https://github.com/Rookie49144/FDT‐Net.

MSTAN: multi-scale spatiotemporal attention network with adaptive relationship mining for remaining useful life prediction in complex systems

Abstract In the era of smart manufacturing and advanced industrial systems, the high degree of integration and intelligence of equipment demands higher reliability and safety from systems. Existing methods often rely on historical data for Remaining Useful Life (RUL) prediction to achieve Prognostic and Health Management (PHM). However, the internal units of complex equipment exhibit significant spatial correlation and temporal diversity, making PHM for complex equipment a multidimensional challenge involving both temporal and spatial information, thereby severely limits the effectiveness of RUL prediction for complex systems. Addressing these challenges, this study introduces a multi-scale spatiotemporal attention network with adaptive relationship mining, specifically designed for the remaining useful life (RUL) prediction of such equipment. The core of the proposed method lies in the multi-scale feature perception module, which adeptly extracts varied scale features from multidimensional sensor data. Following this, an innovative adaptive relationship mining module is integrated to uncover multi-order coupling relationships between diverse sensors, enhancing the model’s predictive accuracy. Furthermore, a spatiotemporal attention module is employed to discern and emphasize crucial spatiotemporal correlations. To validate the effectiveness and superiority of the proposed method, the Commercial Modular Aero-propulsion System Simulation (C-MAPSS) dataset is employed for comprehensive performance evaluation, the IEEE 2012 PHM bearing dataset is also adopted to demonstrate the generalization and robustness of the proposed method. The results not only show a notable improvement over existing methods but also offer a more intuitive understanding through visual representations, marking a significant stride in enhancing the safety and efficiency of complex systems.

Bridging the Digital Skills Gap in Accounting: The Process Mining Audit Professional Curriculum and Badge

SYNOPSIS We designed, implemented, and evaluated a curriculum that trains accounting students and professionals in process mining, an important emerging technology that is not extensively taught in most degree programs. We partnered with a leading provider of process mining software to create the “Process Mining Audit Professional Badge.” The badge’s curriculum combines modules focused on generic process mining skills and modules focused on process mining applications in auditing settings. In 22 months, 1,532 individuals completed the curriculum. Learners were highly satisfied with the training, ranked it among the most effective training modalities they had experienced, and perceived that they learned a great deal about both process mining and the auditing process. We demonstrate an approach to developing materials on emerging technologies that is perceived as useful by learners. If adopted more broadly, the approach could facilitate curricular integration of skills demanded by the profession.

Evidential Multi-Source-Free Unsupervised Domain Adaptation.

Multi-Source-Free Unsupervised Domain Adaptation (MSFUDA) requires aggregating knowledge from multiple source models and adapting it to the target domain. Two challenges remain: 1) suboptimal coarse-grained (domain-level) aggregation of multiple source models, and 2) risky semantics propagation based on local structures. In this article, we propose an evidential learning method for MSFUDA, where we formulate two uncertainties, i.e. Evidential Prediction Uncertainty (EPU) and Evidential Adjacency-Consistent Uncertainty (EAU), respectively for addressing the two challenges. The former, EPU, captures the uncertainty of a sample fitted to a source model, which can suggest the preferences of target samples for different source models. Based on this, we develop an EPU-Based Multi-Source Aggregation module to achieve fine-grained, instance-level source knowledge aggregation. The latter, EAU, provides a robust measure of consistency among adjacent samples in the target domain. Utilizing this, we develop an EAU-Guided Local Structure Mining module to ensure the trustworthy propagation of semantics. The two modules are integrated into the Evidential Aggregation and Adaptation Framework (EAAF), and we demonstrated that this framework achieves state-of-the-art performances on three MSFUDA benchmarks.

VLM-guided Explicit-Implicit Complementary novel class semantic learning for few-shot object detection

Few-shot object detection (FSOD) aims at learning a novel class object detector with abundant base class samples and a limited number of novel class samples. Some recent methods assume that base class images contain unlabeled novel class instances and mine these instances to address the data scarcity of novel classes. Despite achieving competitive results, these methods face two issues: First, mined instances are sampled around the provided few-shot instances, establishing biased data distributions to represent novel classes. Second, the rich novel class semantics contained in base class instances are largely ignored. Benefits from vision-language models (VLMs), we propose a VLM-guided Explicit-Implicit Complementary novel class semantic learning (VEIC) method for FSOD. VEIC incorporates a VLM-cooperated sample mining module that leverages the powerful semantic alignment capability of VLMs to mine representative unlabeled explicit novel instances. In addition, for implicit novel instances (ambiguous base instances contain novel class semantics), we employ a spatial-wise feature re-expression module to capture prominent novel class local patches. Then a dynamic class-transition label assign strategy is proposed to learn from implicit novel instances by utilizing deep network memorization, where implicit novel instances are regarded more as novel classes during the early training and later regarded more as base classes. Experimentally, learning novel class semantics from two complementary types of instances reaches the best performance. Furthermore, learning exclusively from implicit novel instances extends the pseudo-labeling method to a practical scenario where base class images do not contain unlabeled explicit novel instances.

Transformer-based cross-modality interaction guidance network for RGB-T salient object detection

Exploring more effective multimodal fusion strategies is still challenging for RGB-T salient object detection (SOD). Most RGB-T SOD methods tend to focus on the strategy of acquiring modal complementary features by utilizing foreground information while ignoring the importance of background information for salient object localization. In addition, feature fusion without information filtering may introduce more noise. To solve these problems, this paper proposes a new cross-modal interaction guidance network (CIGNet) for RGB-T saliency object detection. Specifically, we construct a transformer-based dual-stream encoder to extract multimodal features. In the decoder, we propose an attention mechanism-based modal information complementary module (MICM) for capturing cross-modal complementary information for global comparison and salient object localization. Based on the MICM features, we design a multi-scale adaptive fusion module (MAFM) to find the optimal salient region of the multi-scale fusion process and reduce redundant features. In order to enhance the completeness of salient features after multi-scale feature fusion, this paper proposes the saliency region mining module (SRMM), which corrects the features in the boundary neighborhood by exploiting the differences between foreground and background pixels and the boundary. Comparisons with other state-of-the-art methods on three RGB-T datasets and five RGB-D datasets, the experimental results demonstrate the superiority and extensiveness of the proposed CIGNet.

Artificial intelligence based social robots in the process of student mental health diagnosis

This paper in order to achieve the application of artificial intelligence based social robots in the process of student mental health diagnosis. When designing the architecture of social robots, factors such as interactivity, adaptability, and scalability were taken into consideration to ensure that they possess human like interaction characteristics and flexibility. Subsequently, a model was constructed based on deep learning technology to achieve functions such as sentiment classification, text mining, and optimization strategies. The input data set of the training model comes from the user’s interaction records and behavior data on the Internet social platform, as well as the user’s feedback information in the process of using the robot. The research on psychological data classification has constructed corresponding algorithms based on pointer networks and text models to achieve text feature extraction and classification. The psychological emotion mining module extracts emotional states from user discourse and maps them to corresponding categories of psychological problems. Finally, based on the user input question content, classify and optimize psychological problems. Research has shown that the robot has certain accuracy and practicality in data classification and student mental health diagnosis.

Alleviating negative group polarization with the aid of social bots

When group polarization is combined with intense negative emotions, it can have detrimental consequences such as online and offline riots. To address this issue, this paper proposes a framework called MNGP (Mitigating Negative Group Polarization) that utilizes both human guidance and social robot assistance. MNGP comprises three main modules: data collection, data mining, and robot embedding. In the data collection module, relevant data capturing group polarization with negative emotions is collected, filtered, and stored. Subsequently, in the data mining module, user and comment characteristics that can help mitigate group polarization are extracted using a novel depolarization index. Finally, in the robot embedding module, a set of social bots are deployed to emulate these characteristics and post neutral or positive comments under manual supervision. Experimental results demonstrate that, over a span of 28 days, with the assistance of 20 social bots across four topics, there were a total of 1039 user interactions and 2349 comments received. Furthermore, the presence of social bots led to a 20 percent reduction in group polarization compared to scenarios where no bots were utilized.

Boundary sample-based class-weighted semi-supervised learning for malignant tumor classification of medical imaging.

Medical image classification plays a pivotal role within the field of medicine. Existing models predominantly rely on supervised learning methods, which necessitate large volumes of labeled data for effective training. However, acquiring and annotating medical image data is both an expensive and time-consuming endeavor. In contrast, semi-supervised learning methods offer a promising approach by harnessing limited labeled data alongside abundant unlabeled data to enhance the performance of medical image classification. Nonetheless, current methods often encounter confirmation bias due to noise inherent in self-generated pseudo-labels and the presence of boundary samples from different classes. To overcome these challenges, this study introduces a novel framework known as boundary sample-based class-weighted semi-supervised learning (BSCSSL) for medical image classification. Our method aims to alleviate the impact of intra- and inter-class boundary samples derived from unlabeled data. Specifically, we address reliable confidential data and inter-class boundary samples separately through the utilization of an inter-class boundary sample mining module. Additionally, we implement an intra-class boundary sample weighting mechanism to extract class-aware features specific to intra-class boundary samples. Rather than discarding such intra-class boundary samples outright, our approach acknowledges their intrinsic value despite the difficulty associated with accurate classification, as they contribute significantly to model prediction. Experimental results on widely recognized medical image datasets demonstrate the superiority of our proposed BSCSSL method over existing semi-supervised learning approaches. By enhancing the accuracy and robustness of medical image classification, our BSCSSL approach yields considerable implications for advancing medical diagnosis and future research endeavors.

Instance-aware diversity feature generation for unsupervised person re-identification

Unsupervised person re-identification (Re-ID) methods have made significant progress by exploiting contrastive learning from unlabeled data. However, the previous approaches including cluster-level or instance-level contrast loss, did not fully explore inherent commonality of each identified individual from unlabeled samples, where the divergence of individual cluster and convergence of different clusters leads to a set of noisy pseudo labels which may result in label noise accumulation. To address this issue, we propose an instance-aware diversity feature generation (IDFG) framework, which can form a stable clustering feature space by exhuming diverse counterparts of given exemplars to update memory dictionary of each cluster, so as to reduce the effect of noisy labels. Specifically, we combines instance segmentation and masked auto-encoder to generate foreground-invariant diversity counterparts of given exemplars to reduce inter-class convergence caused by background similarity between different identification instances. Further, we introduce an instance-aware diversity feature mining module, which gradually creates more reliable clusters to generate more robust pseudo labels by exploiting the compactness and independence of clustering to update the memory dictionary. Extensive experiments demonstrate that the proposed IDFG framework achieves impressive performances of 85.6%, 73.7%, and 31.0% mAP on Market1501, DukeMTMC-reID and MSMT17, respectively.

Enhancing cross-market recommendations by addressing negative transfer and leveraging item co-occurrences

Real-world multinational e-commerce companies, such as Amazon and eBay, serve in multiple countries and regions. Some markets are data-scarce, while others are data-rich. In recent years, cross-market recommendation (CMR) has been proposed to bolster data-scarce markets by leveraging auxiliary information from data-rich markets. Previous CMR algorithms have employed techniques such as sharing market-agnostic parameters or incorporating inter-market similarity to optimize the performance of CMR. However, the existing approaches have several limitations: (1) They do not fully utilize the valuable information on item co-occurrences obtained from data-rich markets (such as the consistent purchase of mice and keyboards). (2) They ignore the issue of negative transfer stemming from disparities across diverse markets. To address these limitations, we introduce a novel attention-based model that exploits users’ historical behaviors to mine general patterns from item co-occurrences and designs market-specific embeddings to mitigate negative transfer. Specifically, we propose an attention-based user interest mining module to harness the potential of common items as bridges for mining general knowledge from item co-occurrence patterns through rich data derived from global markets. In order to mitigate the adverse effects of negative transfer, we decouple the item representations into market-specific embeddings and market-agnostic embeddings. The market-specific embeddings effectively model the inherent biases associated with different markets, while the market-agnostic embeddings learn generic representations of the items. Extensive experiments conducted on seven real-world datasets illustrate our model’s effectiveness.11Our codes and checkpoints are available at https://github.com/laowangzi/ACMR. Our model outperforms the suboptimal model by an average of 4.82%, 6.82%, 3.87%, and 5.34% across four variants of two metrics. Extensive experiments and analysis demonstrate the effectiveness of our proposed model in mining general item co-occurrence patterns and avoiding negative transfer for data-sparse markets.

CEDR: Contrastive Embedding Distribution Refinement for 3D point cloud representation

The distinguishable deep features are essential for the 3D point cloud recognition as they influence the search for the optimal classifier. Most existing point cloud classification methods mainly focus on local information aggregation while ignoring the feature distribution of the whole dataset that indicates more informative and intrinsic semantic relationships of labeled data, if better exploited, which could learn more distinguishing inter-class features. Our work attempts to construct a more distinguishable feature space through performing feature distribution refinement inspired by contrastive learning and sample mining strategies, without modifying the model architecture. To explore the full potential of feature distribution refinement, two modules are involved to boost exceptionally distributed samples distinguishability in an adaptive manner: (i) Confusion-Prone Classes Mining (CPCM) module is aimed at hard-to-distinct classes, which alleviates the massive category-level confusion by generating class-level soft labels; (ii) Entropy-Aware Attention (EAA) mechanism is proposed to remove influence of the trivial cases which could substantially weaken model performance. Our method achieves competitive results on multiple applications of point cloud. In particular, our method gets 85.8% accuracy on ScanObjectNN, and substantial performance gains up to 2.7% in DCGNN, 3.1% in PointNet++, and 2.4% in GBNet. Our code is available at https://github.com/YangFengSEU/CEDR.

English Teaching Quality Evaluation Model for Higher Education Based on Educational Data Mining Technology

Constructing teaching quality evaluation systems for higher vocational education is beneficial for schools, serving as the cornerstone for developing abilities and enabling precise education management. In this publication, a Dual Relational Graph Attention Network (ETQE-DRGAN)-Based Educational Data Mining Technology English Teaching Quality Evaluation Model for Higher Education is proposed. In three ways, this study assesses and defends the system. An overview of the system's primary components, such as the user and data administration modules, data mining, online assessment, and inquiry modules, is given in the first section. Frequent item sets are divided into two stages and the development of invalid candidate sets is limited to avoid the formation of negative association rules. The Tid-list Vertical Partitioning Query-Apriori Algorithm (TPQ-Apriori algorithm) is improved by these two approaches. The system analysis is covered in section three. In order to assess the success of instruction, this work incorporates a Dual Relational Graph Attention Networks method that takes into account several factors such as age, gender, education, teaching attitude, teaching topic, teaching technique, and teaching effectiveness. The proposed ETQE-DRGAN is implemented in MATLAB, using the School Database for calculation. Numerous performance metrics like response time, precision, accuracy, sensitivity, and F1-score are utilized to calculate the effectiveness of ETQE-DRGAN technique. The obtained outcomes indicate that the ETQE-DRGAN approach achieves the highest precision of 98.5%, Accuracy of 98% and fastest response time of 6.95swhen compared to existing methods like ETQE-AA, ETQE-DTA and ETQE-ML.

Polyp segmentation with interference filtering and dynamic uncertainty mining.

Objective.Accurate polyp segmentation from colo-noscopy images plays a crucial role in the early diagnosis and treatment of colorectal cancer. However, existing polyp segmentation methods are inevitably affected by various image noises, such as reflections, motion blur, and feces, which significantly affect the performance and generalization of the model. In addition, coupled with ambiguous boundaries between polyps and surrounding tissue, i.e. small inter-class differences, accurate polyp segmentation remains a challenging problem.Approach.To address these issues, we propose a novel two-stage polyp segmentation method that leverages a preprocessing sub-network (Pre-Net) and a dynamic uncertainty mining network (DUMNet) to improve the accuracy of polyp segmentation. Pre-Net identifies and filters out interference regions before feeding the colonoscopy images to the polyp segmentation network DUMNet. Considering the confusing polyp boundaries, DUMNet employs the uncertainty mining module (UMM) to dynamically focus on foreground, background, and uncertain regions based on different pixel confidences. UMM helps to mine and enhance more detailed context, leading to coarse-to-fine polyp segmentation and precise localization of polyp regions.Main results.We conduct experiments on five popular polyp segmentation benchmarks: ETIS, CVC-ClinicDB, CVC-ColonDB, EndoScene, and Kvasir. Our method achieves state-of-the-art performance. Furthermore, the proposed Pre-Net has strong portability and can improve the accuracy of existing polyp segmentation models.Significance.The proposed method improves polyp segmentation performance by eliminating interference and mining uncertain regions. This aids doctors in making precise and reduces the risk of colorectal cancer. Our code will be released athttps://github.com/zyh5119232/DUMNet.