In the context of neuro-pathological disorders, neuroimaging has been widely accepted as a clinical tool for diagnosing patients with Alzheimer’s disease (AD) and mild cognitive impairment (MCI). The advanced deep learning method, novel brain imaging technique, was applied in this study to evaluate its contribution improving diagnostic accuracy AD. Three-dimensional convolutional neural networks (3D-CNNs) were magnetic resonance (MRI) execute binary ternary classification models. dataset...

Regular crack detection is essential for extending the service life of bridges. However, image data collected during bridge inspections are complex to convert into physical information and construct intuitive comprehensive Three-Dimensional (3D) models incorporating information. An intelligent method surface damage based on Unmanned Aerial Vehicles (UAVs) proposed these challenges, a three-stage detection, quantification, visualization process. This enables automatic localization in 3D...

Pulmonary lobe segmentation is an important task for pulmonary disease related Computer Aided Diagnosis systems (CADs). Classical methods rely on successful detection of fissures and other anatomical information such as the location blood vessels airways. With success deep learning in recent years, Deep Convolutional Neural Network (DCNN) has been widely applied to analyze medical images like Computed Tomography (CT) Magnetic Resonance Imaging (MRI), which, however, requires a large number...

The issues of both system security and safety can be dissected integrally from the perspective behavioral appropriateness . That is, a that is secure or safe judged by whether behavior certain agent(s) appropriate not. Specifically, so-called involves right agent performing actions at time under conditions. Then, according to different levels degrees custodies, authentication graded into three levels, i.e. , Identity Conformity Benignity In broad sense, for issue, not only an innovative...

Segment Anything Model 2 (SAM 2), a prompt-driven foundation model extending SAM to both image and video domains, has shown superior zero-shot performance compared its predecessor. Building on SAM's success in medical segmentation, presents significant potential for further advancement. However, similar SAM, is limited by output of binary masks, inability infer semantic labels, dependence precise prompts the target object area. Additionally, direct application segmentation tasks yields...

Bayesian network is an effective uncertain knowledge representation and reasoning method. Fuzzy sets can be used for expressing fuzzy events or objectives in some special region. Combining these two theories, this paper discusses the probability of event presents a hybrid inference system with networks which are called "Fuzzy Networks (FBNs) ". Then case demonstrates validity FBNs machinery fault diagnosis.

Advances in self-supervised learning have drawn attention to developing techniques extract effective visual representations from unlabeled images. Contrastive (CL) trains a model consistent features by generating different views. Recent success of Masked Autoencoders (MAE) highlights the benefit generative modeling learning. The approaches encode input into compact embedding and empower model's ability recovering original input. However, our experiments, we found vanilla MAE mainly recovers...

Graph neural networks (GNNs) are playing exciting roles in the application scenarios where features hidden information associations. Fraud prediction of online credit loan services (OCLSs) is such a typical scenario. But it has another rather critical challenge, i.e., scarcity data labels. Fortunately, GNNs can also cope with this problem due to their good ability semi-supervised learning by mining structure and feature within graphs. Nevertheless, gain internal often too limited help handle...

Centralized training and decentralized execution (CTDE) paradigm is widely employed to address the nonstationary partial observability in multiagent reinforcement learning (MARL). One of main challenges that restricts performance CTDE <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">credit assignment.</i> Existing methods cannot sufficiently energize each agent for exploring a broader solution space without compromising or factorization...

According to the mass health monitoring data accumulated of bridge structure for a long time, this paper proposes method reconstitute time series in phase space. Since points are regressed by support vector machine (SVM), relevant past behavior patterns established. Then, it could infer future development trend and form basis online security early warning structure. The strain tilt Pian Yan-zi Chongqing analyzed compared with prediction auto regression moving average (ARMA). results show...

Root cause analysis (RCA) of network faults is crucial to wireless operation and management. It, however, challenging, due diverse feature types, lengths time slices, simultaneous occurrences multiple root causes, lack training samples. In this paper, we present our solutions for these problems in ICASSP-SPGC-2022 AIOps Challenge Communication Networks. We first design specific engineering method represent the provided spatial features. Secondly, conduct series on data propose an efficient...

Anti-fraud engineering for online credit loan (OCL) platforms is getting more challenging due to the developing specialization of gang fraud. Associations are critical features referring assessing credibility applications OCL fraud prediction. State-of-the-art solutions employ graph-based methods mine hidden associations among effectively. They perform well based on information asymmetry which guaranteed by huge advantage over fraudsters in terms data quantity and quality at their disposal....

In order to make full use of fault historical text data, knowledge graph is used assist diagnosis. Text classification, a fundamental method for constructing graph, prone the overfitting issue when dealing with sparse and small sample data such as domain data. Therefore, an integrated neural network model based on CNN-BiLSTM (convolutional networks bidirectional long short-term memory networks) proposed in this paper. model, CNN layers extract local semantic features while BiLSTM integrate...

RGB-Thermal Salient Object Detection aims to pinpoint prominent objects within aligned pairs of visible and thermal infrared images. Traditional encoder-decoder architectures, while designed for cross-modality feature interactions, may not have adequately considered the robustness against noise originating from defective modalities. Inspired by hierarchical human visual systems, we propose ConTriNet, a robust Confluent Triple-Flow Network employing Divide-and-Conquer strategy. Specifically,...

Pulmonary lobe segmentation is an important task for pulmonary disease related Computer Aided Diagnosis systems (CADs). Classical methods rely on successful detection of fissures and other anatomical information such as the location blood vessels airways. With success deep learning in recent years, Deep Convolutional Neural Network (DCNN) has been widely applied to analyze medical images like Computed Tomography (CT) Magnetic Resonance Imaging (MRI), which, however, requires a large number...

The issues of both system security and safety can be dissected integrally from the perspective behavioral \emph{appropriateness}. That is, a is secure or safe judged by whether behavior certain agent(s) \emph{appropriate} not. Specifically, so-called \emph{appropriate behavior} involves right agent performing actions at time under conditions. Then, according to different levels appropriateness degrees custodies, authentication graded into three levels, i.e., \emph{Identity},...

There are a large amount of complex data in the power grid database. How to use these reasonably and efficiently solve problem operation safety has become an urgent need. Based on principle association rule mining technology, this paper realizes potential information massive database through improved Apriori algorithm., The algorithm innovation introduces two factors efficiency interest, so that strong rules can be screened more accurately. In addition, view number candidate sets algorithm,...