We study algorithmic strategies to effectively contain diffusion via edge immunization. We consider the scenarios where the epidemic characteristics are known and unknown, and accordingly present approaches relying upon the epidemic dynamics and the network topology. In particular, for the former, we propose the greedy and inverse greedy immune schemes to greedily minimize the immune edge set, whose performance is guaranteed by the exhaustive search. We also present strategies to scale up the associated methods such that they can more efficiently obtain better solutions. For the latter, we find that the epidemic incidence of the immunized network is determined by both local and global connection patterns, characterized by the critical threshold in network epidemiology and the largest connected component in network percolation, respectively. Thus, we propose network topology-based strategies that obtain the immune edge set by simultaneously minimizing both the factors. We conduct extensive experiments on synthetic and empirical social networks to evaluate the proposed methods. Results show that our methods outperform the state-of-the-art by a large margin. Besides, the developed topology-based strategies can always obtain comparable results to the greedy strategies, which are computationally much cheaper than existing approaches and thus favorable to tackle large-scale networks.

This paper studies the network dismantling (ND) problem and aims to develop more effective models and approaches to cope with it, such that a given network can be dismantled by a set of vital nodes of minimum size. To achieve that, we propose a three-phase framework—the Percolation coarsening, Belief propagation dismantling, and Fragmentation optimization fine-tuning (PBF) framework—consisting of PBF-I, PBF-II, and PBF-III, where we contribute three new and one improved algorithms. In particular, PBF-I studies strategies to effectively coarsen the studied network via the merger of less influential nodes, such that the computational efficiency of the follow-up PBF-II phase can be maximized. PBF-II considers the superiority of the belief propagation (BP) algorithm in the ND problem and proposes an improved BP to identify vital nodes from the coarse-grained network, which particularly focuses on the largest connected component and obtains the vital nodes from a filtered candidate set. In addition, PBF-III presents fine-tuning strategies to further improve the quality of solutions obtained in PBF-II. The effectiveness of the proposed framework is validated on over 10 empirical networks in regard to varied circumstances. Our results show that the developed framework can obtain dismantling node sets of much smaller sizes compared to the state-of-the-art in almost all cases. Meanwhile, our framework is also more effective, efficient, and stable compared to existing methods, and is capable of tackling the ND problem in extremely large networks. We are convinced that the model and methodology introduced in this paper could be applied to many applications, such as the robustness and resilience analysis of network-structural infrastructures, the suppression of epidemics, and the containment of misinformation on social networks. The source code of the proposed PBF framework will be made publicly available upon acceptance of the manuscript.

Missing modalities pose a major issue in Alzheimer's Disease (AD) diagnosis, as many subjects lack full imaging data due to cost and clinical constraints. While multi-modal learning leverages complementary information, most existing methods train only on complete data, ignoring the large proportion of incomplete samples in real-world datasets like ADNI. This reduces the effective training set and limits the full use of valuable medical data. While some methods incorporate incomplete samples, they fail to effectively address inter-modal feature alignment and knowledge transfer challenges under high missing rates. To address this, we propose a Prototype-Guided Adaptive Distillation (PGAD) framework that directly incorporates incomplete multimodal data into training. PGAD enhances missing modality representations through prototype matching and balances learning with a dynamic sampling strategy. We validate PGAD on the ADNI dataset with varying missing rates$(20 \%, 50 \%$, and 70 %) and demonstrate that it significantly outperforms state-of-the-art approaches. Ablation studies confirm the effectiveness of prototype matching and adaptive sampling, highlighting the potential of our framework for robust and scalable AD diagnosis in real-world clinical settings.

Glaucoma is one of the major eye diseases that leads to progressive optic nerve fiber damage and irreversible blindness, afflicting millions of individuals. Glaucoma forecast is a good solution to early screening and intervention of potential patients, which is helpful to prevent further deterioration of the disease. It leverages a series of historical fundus images of an eye and forecasts the likelihood of glaucoma occurrence in the future. However, the irregular sampling nature and the imbalanced class distribution are two challenges in the development of disease forecasting approaches. To this end, we introduce the Multi-scale Spatio-temporal Transformer Network (MST-former) based on the transformer architecture tailored for sequential image inputs, which can effectively learn representative semantic information from sequential images on both temporal and spatial dimensions. Specifically, we employ a multi-scale structure to extract features at various resolutions, which can largely exploit rich spatial information encoded in each image. Besides, we design a time distance matrix to scale time attention in a non-linear manner, which could effectively deal with the irregularly sampled data. Furthermore, we introduce a temperature-controlled Balanced Softmax Cross-entropy loss to address the class imbalance issue. Extensive experiments on the Sequential fundus Images for Glaucoma Forecast (SIGF) dataset demonstrate the superiority of the proposed MST-former method, achieving an AUC of 96.6% for glaucoma forecasting. Besides, our method shows excellent generalization capability on the Alzheimer's Disease Neuroimaging Initiative (ADNI) MRI dataset, with an accuracy of 88.2% for mild cognitive impairment and Alzheimer's disease prediction, outperforming the compared method by a large margin. A series of ablation studies further verify the contribution of our proposed components in addressing the irregular sampled and class imbalanced problems.

Skin lesion segmentation is vital for the early detection, diagnosis, and treatment of melanoma, yet it remains challenging due to significant variations in lesion attributes (e.g., color, size, shape), ambiguous boundaries, and noise interference. Recent advancements have focused on capturing contextual information and incorporating boundary priors to handle challenging lesions. However, there has been limited exploration on the explicit analysis of the inherent patterns of skin lesions, a crucial aspect of the knowledge-driven decision-making process used by clinical experts. In this work, we introduce a novel approach called Probabilistic Attribute Learning (PAL), which leverages knowledge of lesion patterns to achieve enhanced performance on challenging lesions. Recognizing that the lesion patterns exhibited in each image can be properly depicted by disentangled attributes, we begin by explicitly estimating the distributions of these attributes as distinct Gaussian distributions, with mean and variance indicating the most likely pattern of that attribute and its variation. Using Monte Carlo Sampling, we iteratively draw multiple samples from these distributions to capture various potential patterns for each attribute. These samples are then merged through an effective attribute fusion technique, resulting in diverse representations that comprehensively depict the lesion class. By performing pixel-class proximity matching between each pixel-wise representation and the diverse class-wise representations, we significantly enhance the model's robustness. Extensive experiments on two public skin lesion datasets and one unified polyp lesion dataset demonstrate the effectiveness and strong generalization ability of our method. Codes are available at https://github.com/IsYuchenYuan/PAL.

Label scarcity, class imbalance and data uncertainty are three primary challenges that are commonly encountered in the semi-supervised medical image segmentation. In this work, we focus on the data uncertainty issue that is overlooked by previous literature. To address this issue, we propose a probabilistic prototype-based classifier that introduces uncertainty estimation into the entire pixel classification process, including probabilistic representation formulation, probabilistic pixel-prototype proximity matching, and distribution prototype update, leveraging principles from probability theory. By explicitly modeling data uncertainty at the pixel level, model robustness of our proposed framework to tricky pixels, such as ambiguous boundaries and noises, is greatly enhanced when compared to its deterministic counterpart and other uncertainty-aware strategy. Empirical evaluations on three publicly available datasets that exhibit severe boundary ambiguity show the superiority of our method over several competitors. Moreover, our method also demonstrates a stronger model robustness to simulated noisy data. Code is available at https://github.com/IsYuchenYuan/PPC.

This article studies the diffusion-source-inference (DSI) problem, whose solution plays an important role in real-world scenarios such as combating misinformation and controlling diffusions of information or disease. The main task of the DSI problem is to optimize an estimator, such that the real source can be more precisely targeted. In this article, we assume that the state of a number of nodes, called observer set, in a network could be investigated if necessary, and study what configuration of those nodes could facilitate a better solution for the DSI problem. In particular, we find that the conventional error distance metric cannot precisely evaluate the effectiveness of varied DSI approaches in heterogeneous networks, and thus propose a novel and more general measurement, the candidate set, that is formulated to contain the diffusion source for sure. We propose the percolation-based evolutionary framework (PrEF) to optimize the observer set such that the candidate set can be minimized. Hence, one could further conduct more intensive investigation or search on only a few nodes to target the source. To achieve that, we first theoretically show that the size of the candidate set is bounded by the size of the largest component cover, and demonstrate that there are some similarities between the DSI problem and the network immunization problem. We find that, given the associated direction information of the diffusion is known on observers, the minimization of the candidate set is equivalent to the minimization of the order parameter if we view the observer set as the removal node set. Hence, PrEF is developed based on the network percolation and evolutionary algorithm. The effectiveness of the proposed method is validated on both synthetic and empirical networks in regard to varied circumstances. Our results show that the developed approach could achieve much smaller candidate sets compared to the state of the art in almost all cases, e.g., it is better in 26 out of 27 empirical networks and 155 out of 162 cases regarding the critical threshold. Meanwhile, our approach is also more stable, i.e., it works well irrespective of varied infection probabilities, diffusion models, and underlying networks. More importantly, we provide a framework for the analysis of the DSI problem in large-scale networks.

A great number of studies have demonstrated that many complex systems could benefit a lot from complex networks, through either a direct modeling on which dynamics among agents could be investigated in a global view or an indirect representation by the aid of that the leading factors could be captured more clearly. Hence, in the context of networks, this article copes with the continuous network dismantling problem which aims to find the key node set whose removal would break down a given network more thoroughly and thus is more capable of suppressing virus or misinformation. To achieve this goal effectively and efficiently, we propose the external-degree and internal-size component suppression (EDIS) framework based on the network percolation, where we constrain the search space by a well-designed local goal function and candidate selection approach such that EDIS could obtain better results than the-state-of-the-art in networks of millions of nodes in seconds. We also contribute two strategies with time complexity${\mathcal {O}}(m\log _{\vartheta } m)$and space complexity${\mathcal {O}}(m)$, of networks of m edges, under such framework by well studying the evolving characteristics of the associated connected components as nodes are occupied, where$\vartheta \gt 1$is a hyperparameter. Our results on 12 empirical networks from various domains demonstrate that the proposed method has far better performance than the-state-of-the-art over both effectiveness and computing time. Our study could play important roles in many real-world scenarios, such as the containment of misinformation or epidemics, the distribution of resources or vaccine, the decision of which group of individuals set to quarantine, or the detection of the resilience of a network-based system under intentional attacks.

Multi-modal brain tumor segmentation typically involves four magnetic resonance imaging (MRI) modalities, while incomplete modalities significantly degrade performance. Existing solutions employ explicit or implicit modality adaptation, aligning features across modalities or learning a fused feature robust to modality incompleteness. They share a common goal of encouraging each modality to express both itself and the others. However, the two expression abilities are entangled as a whole in a seamless feature space, resulting in prohibitive learning burdens. In this paper, we propose DeMoSeg to enhance the modality adaptation by Decoupling the task of representing the ego and other Modalities for robust incomplete multi-modal Segmentation. The decoupling is super lightweight by simply using two convolutions to map each modality onto four feature sub-spaces. The first sub-space expresses itself (Self-feature), while the remaining sub-spaces substitute for other modalities (Mutual-features). The Self- and Mutual-features interactively guide each other through a carefully-designed Channel-wised Sparse Self-Attention (CSSA). After that, a Radiologist-mimic Cross-modality expression Relationships (RCR) is introduced to have available modalities provide Self-feature and also ‘lend’ their Mutual-features to compensate for the absent ones by exploiting the clinical prior knowledge. The benchmark results on BraTS2020, BraTS2018 and BraTS2015 verify the DeMoSeg’s superiority thanks to the alleviated modality adaptation difficulty. Concretely, for BraTS2020, DeMoSeg increases Dice by at least 0.92%, 2.95% and 4.95% on whole tumor, tumor core and enhanced tumor regions, respectively, compared to other state-of-the-arts. Codes are at https://github.com/kk42yy/DeMoSeg.

The multi-view hash method is crucial in multimedia retrieval via transforming heterogeneous data from multiple views into binary hash codes. Existing methods primarily focus on leveraging complementary information across multiple views, while ignoring the issue of imbalanced optimization. That is, the features from some views in multi-view data are stronger than others, which leads to a less optimization of the networks handling those weaker features. To fully utilize the data from all the views, we propose a novel Adaptive Loss-aware Modulation (ALM) method to address this imbalance issue during the fusion of multi-view features. Specifically, in training, ALM automatically calculates the total loss for each view to reflect the performance of the respective view's backbone network. The modulation coefficient is then determined based on the total loss of the corresponding view. By multiplying the gradient of the network of each view with its corresponding modulation coefficient, we can suppress the gradient update rate of the view with stronger features, while maintaining the normal gradient update rate for the ones with weaker features. Based on ALM, we further introduce a new Balanced Multi-View Hashing (BMVH) method. Extensive experiments on three public datasets demonstrate that the proposed BMVH outperforms state-of-the-art methods, with a maximum increase of 3.22% in mAP.

We study the containment of diffusion in a network immunization perspective, whose solution also plays fundamental roles in scenarios such as the inference of rumor sources and the control of malicious viral marketings. In general, the network immunization aims to suppress the giant connected component of a network by removing as fewer nodes as possible, so that the intervention of the transmission could be achieved by only a few resources. Here, rather than that and based on the fact that removing edges might be cheaper and more applicable in some scenarios, we investigate which group of edges whose removal could boost the performance of an immunization strategy more effectively. We consider both cases that the network topology is known and unknown, and thus two approaches are accordingly developed based on the Edge RelationShip (ERS) and Explosive Percolation over Partial (EPP) information. We evaluate the performance of ERS by comparing it with strategies based on the edge betweenness, the product of eigenvector centralities of the nodes connected by edges, the epidemic link equations, etc. Results on over 30 real networks show that ERS could effectively acquire far better solutions by much less computing time. We also demonstrate the performance of EPP in the circumstances of decentralized, centralized, and delayed cases. We find that the performance of EPP would be in a degree degraded by the uncertainty of inferences from individuals, inaccuracy of predictions, and delay of reactions. But in almost all cases, the developed approach can more effectively suppress a diffusion compared to the currently random strategy, especially when a tough restriction is needed or a combination with the acquaintance immunization is conducted.

This paper addresses the Fast outbreak Sensing and Effective diffusion source Inferring (FSEI) problem, which assumes that the state of nodes in a particularly chosen observer set can be monitored if necessary and aims to optimize the observer set such that outbreaks can be timely detected and their sources can be effectively targeted. We propose three approaches to tackle the FSEI problem: Greedy Strategy (GS), Network-Topology-based Method (NTM), and Hybrid Method (HM). Among them, GS relies on collected outbreaks and constructs the observer set by iteratively choosing and removing the node that minimizes the product of sensing time and source targeting cost of the remaining network. For NTM, we also consider the remaining network and introduce a novel strategy to optimize its topology via simultaneously minimizing the adjoining component size and ratio of the first and second moments. HM is a combination of GS and NTM, considering the submodular property of GS on the minimization of the sensing time and well approximation of the component size on the optimization of the source targeting. We perform extensive experiments on over 200 empirical networks, using various diffusion models, to validate the proposed methods. The results demonstrate that our approaches consistently outperform the state-of-the-art. We believe that the model and methodology presented in this paper can be readily applied to real-world scenarios such as combating misinformation and controlling diffusions of information or disease.

Histopathology image analysis serves as the gold standard for cancer diagnosis. Efficient and precise diagnosis is quite critical for the subsequent therapeutic treatment of patients. So far, computer-aided diagnosis has not been widely applied in pathological field yet as currently well-addressed tasks are only the tip of the iceberg. Whole slide image (WSI) classification is a quite challenging problem. First, the scarcity of annotations heavily impedes the pace of developing effective approaches. Pixelwise delineated annotations on WSIs are time consuming and tedious, which poses difficulties in building a large-scale training dataset. In addition, a variety of heterogeneous patterns of tumor existing in high magnification field are actually the major obstacle. Furthermore, a gigapixel scale WSI cannot be directly analyzed due to the immeasurable computational cost. How to design the weakly supervised learning methods to maximize the use of available WSI-level labels that can be readily obtained in clinical practice is quite appealing. To overcome these challenges, we present a weakly supervised approach in this article for fast and effective classification on the whole slide lung cancer images. Our method first takes advantage of a patch-based fully convolutional network (FCN) to retrieve discriminative blocks and provides representative deep features with high efficiency. Then, different context-aware block selection and feature aggregation strategies are explored to generate globally holistic WSI descriptor which is ultimately fed into a random forest (RF) classifier for the image-level prediction. To the best of our knowledge, this is the first study to exploit the potential of image-level labels along with some coarse annotations for weakly supervised learning. A large-scale lung cancer WSI dataset is constructed in this article for evaluation, which validates the effectiveness and feasibility of the proposed method. Extensive experiments demonstrate the superior performance of our method that surpasses the state-of-the-art approaches by a significant margin with an accuracy of 97.3%. In addition, our method also achieves the best performance on the public lung cancer WSIs dataset from The Cancer Genome Atlas (TCGA). We highlight that a small number of coarse annotations can contribute to further accuracy improvement. We believe that weakly supervised learning methods have great potential to assist pathologists in histology image diagnosis in the near future.

Deep learning has achieved remarkable success in the optical coherence tomography (OCT) image classification task with substantial labelled B-scan images available. However, obtaining such fine-grained expert annotations is usually quite difficult and expensive. How to leverage the volume-level labels to develop a robust classifier is very appealing. In this paper, we propose a weakly supervised deep learning framework with uncertainty estimation to address the macula-related disease classification problem from OCT images with the only volume-level label being available. First, a convolutional neural network (CNN) based instance-level classifier is iteratively refined by using the proposed uncertainty-driven deep multiple instance learning scheme. To our best knowledge, we are the first to incorporate the uncertainty evaluation mechanism into multiple instance learning (MIL) for training a robust instance classifier. The classifier is able to detect suspicious abnormal instances and abstract the corresponding deep embedding with high representation capability simultaneously. Second, a recurrent neural network (RNN) takes instance features from the same bag as input and generates the final bag-level prediction by considering the individually local instance information and globally aggregated bag-level representation. For more comprehensive validation, we built two large diabetic macular edema (DME) OCT datasets from different devices and imaging protocols to evaluate the efficacy of our method, which are composed of 30,151 B-scans in 1,396 volumes from 274 patients (Heidelberg-DME dataset) and 38,976 B-scans in 3,248 volumes from 490 patients (Triton-DME dataset), respectively. We compare the proposed method with the state-of-the-art approaches, and experimentally demonstrate that our method is superior to alternative methods, achieving volume-level accuracy, F1-score and area under the receiver operating characteristic curve (AUC) of 95.1%, 0.939 and 0.990 on Heidelberg-DME and those of 95.1%, 0.935 and 0.986 on Triton-DME, respectively. Furthermore, the proposed method also yields competitive results on another public age-related macular degeneration OCT dataset, indicating the high potential as an effective screening tool in the clinical practice.

Deep learning approaches have demonstrated remarkable progress in automatic Chest X-ray analysis. The data-driven feature of deep models requires training data to cover a large distribution. Therefore, it is substantial to integrate knowledge from multiple datasets, especially for medical images. However, learning a disease classification model with extra Chest X-ray (CXR) data is yet challenging. Recent researches have demonstrated that performance bottleneck exists in joint training on different CXR datasets, and few made efforts to address the obstacle. In this paper, we argue that incorporating an external CXR dataset leads to imperfect training data, which raises the challenges. Specifically, the imperfect data is in two folds: domain discrepancy, as the image appearances vary across datasets; and label discrepancy, as different datasets are partially labeled. To this end, we formulate the multi-label thoracic disease classification problem as weighted independent binary tasks according to the categories. For common categories shared across domains, we adopt task-specific adversarial training to alleviate the feature differences. For categories existing in a single dataset, we present uncertainty-aware temporal ensembling of model predictions to mine the information from the missing labels further. In this way, our framework simultaneously models and tackles the domain and label discrepancies, enabling superior knowledge mining ability. We conduct extensive experiments on three datasets with more than 360,000 Chest X-ray images. Our method outperforms other competing models and sets state-of-the-art performance on the official NIH test set with 0.8349 AUC, demonstrating its effectiveness of utilizing the external dataset to improve the internal classification.

There are many target methods that are efficient to tackle the robustness and immunization problem, in particular, to identify the most influential nodes in a certain complex network. Unfortunately, owing to the diversity of networks, none of them could be accounted as a universal approach that works well in a wide variety of networks. Hence, in this paper, from a percolation perspective, we connect the immunization and robustness problem with an evolutionary algorithm, i.e., a framework of an evolutionary algorithm for investigation of influential nodes in complex networks, in which we have developed procedures of selection, mutation, and initialization of population as well as maintaining the diversity of population. To validate the performance of the proposed framework, we conduct intensive experiments on a large number of networks and compare it to several state-of-the-art strategies. The results demonstrate that the proposed method has significant advantages over others, especially on empirical networks in most of which our method has over 10% advantages of both optimal immunization threshold and average giant fraction, even against the most excellent existing strategies. Additionally, our discussion reveals that there might be better solutions with various initial methods.

The number of mitoses per tissue area gives an important aggressiveness indication of the invasive breast carcinoma.However, automatic mitosis detection in histology images remains a challenging problem. Traditional methods either employ hand-crafted features to discriminate mitoses from other cells or construct a pixel-wise classifier to label every pixel in a sliding window way. While the former suffers from the large shape variation of mitoses and the existence of many mimics with similar appearance, the slow speed of the later prohibits its use in clinical practice.In order to overcome these shortcomings, we propose a fast and accurate method to detect mitosis by designing a novel deep cascaded convolutional neural network, which is composed of two components. First, by leveraging the fully convolutional neural network, we propose a coarse retrieval model to identify and locate the candidates of mitosis while preserving a high sensitivity.Based on these candidates, a fine discrimination model utilizing knowledge transferred from cross-domain is developed to further single out mitoses from hard mimics.Our approach outperformed other methods by a large margin in 2014 ICPR MITOS-ATYPIA challenge in terms of detection accuracy. When compared with the state-of-the-art methods on the 2012 ICPR MITOSIS data (a smaller and less challenging dataset), our method achieved comparable or better results with a roughly 60 times faster speed.