检索结果-内蒙古大学图书馆

23rd Mexican International Conference on Artificial Intelligence

作者： Martinez, Sergio Salas, Brenda Perez, Nora Neme, Antonio Univ Nacl Autonoma Mexico Postgrad Program Comp Sci Mexico City Mexico Univ Nacl Autonoma Mexico Inst Invest Matemat Aplicadas & Sistemas Mexico City Mexico

ISBN: (纸本)9783031755422;9783031755439

Genomes are complex biological structures that encode information that can be translated onto several levels, such as genes and proteins. Identification of relevant patterns in genomes is of paramount importance, as they may indicate states of biological or medical relevance. Among the patterns that can be detected, anomalies are especially relevant. Anomalies are instances that do not resemble, under certain metrics, the rest of the observations under study. Anomalies and their detection are relevant since their presence may indicate a systematic error in some stage of the analyzed process or structure, or may indicate that the studied system or phenomenon is undergoing a phase transition or other relevant drift in its dynamics. Here, we applied unsupervised anomaly detection algorithms to the codon usage of the genomes of thousands of SARS COV2 virus isolated in Mexico. Codon usage condenses the relative frequency of appearance of nucleotide triplets, or codons, which code for amino acids, the basic blocks of proteins. By applying several algorithms, we detected patterns that are of epidemiological relevance. The detected patterns are anomalous genomes based on their codon usage. Anomalous patterns are relevant not only because they have not been previously detected in data from Mexico, but also because they allow identification of one of the possible sources of the anomalies. Most of these anomalies were identified in two neighboring states in Mexico, namely Puebla and Tlaxcala. In addition, we identified that almost all anomalies come from subjects who were treated in the same laboratory. Based on the evidence we present here, we conclude that anomaly detection algorithms are relevant in the surveillance of epidemics.

关键词： unsupervised anomaly detection codon usage SARS COV2

来源：评论

学校读者我要写书评

暂无评论

GAN-based synthetic FDG PET images from T1 brain MRI can serve to improve performance of deep unsupervised anomaly detection models

引用

COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE 2025年 265卷 108727-108727页

作者： Zotova, Daria Pinon, Nicolas Trombetta, Robin Bouet, Romain Jung, Julien Lartizien, Carole Univ Claude Bernard Lyon 1 INSA Lyon CNRS INSERMCREATIS UMR 5220U1294 F-69621 Lyon France Univ Lyon 1 Lyon Neurosci Res Ctr INSERM U1028 CNRS UMR5292 F-69500 Bron France

Background and Objective: Research in the cross-modal medical image translation domain has been very productive over the past few years in tackling the scarce availability of large curated multi-modality datasets with the promising performance of GAN-based architectures. However, only a few of these studies assessed task-based related performance of these synthetic data, especially for the training of deep models. Methods: We design and compare different GAN-based frameworks for generating synthetic brain [18F]fluorodeoxyglucose (FDG) PET images from T1 weighted MRI data. We first perform standard qualitative and quantitative visual quality evaluation. Then, we explore further impact of using these fake PET data in the training of a deep unsupervised anomaly detection (UAD) model designed to detect subtle epilepsy lesions in T1 MRI and FDG PET images. We introduce novel diagnostic task-oriented quality metrics of the synthetic FDG PET data tailored to our unsupervised detection task, then use these fake data to train a use case UAD model combining a deep representation learning based on siamese autoencoders with a OC-SVM density support estimation model. This model is trained on normal subjects only and allows the detection of any variation from the pattern of the normal population. We compare the detection performance of models trained on 35 paired real MR T1 of normal subjects paired either on 35 true PET images or on 35 synthetic PET images generated from the best performing generative models. Performance analysis is conducted on 17 exams of epilepsy patients undergoing surgery. Results: The best performing GAN-based models allow generating realistic fake PET images of control subject with SSIM and PSNR values around 0.9 and 23.8, respectively and in distribution (ID) with regard to the true control dataset. The best UAD model trained on these synthetic normative PET data allows reaching 74% sensitivity. Conclusion: Our results confirm that GAN-based models are t

关键词： Medical image synthesis Cycle-GAN PET MRI unsupervised anomaly detection Out-of-distribution (OOD) Epileptogenic zone detection

来源：评论

学校读者我要写书评

暂无评论

unsupervised Video anomaly detection Using Video Vision Transformer and Adversarial Training

引用

IEEE ACCESS 2025年 13卷 56169-56179页

作者： Kobayashi, Shimpei Hizukuri, Akiyoshi Nakayama, Ryohei Ritsumeikan Univ Grad Sch Sci & Engn Kusatsu Shiga 5258577 Japan

Surveillance cameras have been recently introduced in various locations to maintain public safety. However, it is tedious for security personnel to continue observing videos obtained by surveillance cameras because abnormal events rarely occur. Consequently, this study aims to develop an unsupervised anomaly detection method for surveillance videos based on the prediction of future frame and optical flow magnitude using adversarial networks. Our database consists of two public datasets for anomaly detection: UCSD Pedestrian 2 (UCSD Ped2) and ShanghaiTech Campus. The proposed network consists of a generator and a U-Net-based discriminator. The generator has an encoder based on a video vision transformer and two decoders that independently predict the future frame and optical flow magnitude using feature maps extracted from the encoder at different resolutions. The U-Net-based discriminator identifies at the pixel level whether the image is a "real" or "false" frame predicted by the generator. The generator and U-Net-based discriminator in the proposed network are trained through adversarial training. Finally, anomaly detection is performed based on element-wise subtraction between the predicted future frame and the optical flow magnitude and their corresponding ground truth frames. The areas under the receiver operating characteristic curve (AUC) with the proposed network were 97.5% for UCSD Ped2 and 85.8% for ShanghaiTech Campus. The AUC for the proposed network was also greater than those of well-known anomaly detection methods.

关键词： anomaly detection Transformers Optical flow Surveillance Generators Cameras Training Kernel Decoding Feature extraction Generative adversarial networks surveillance videos U-Net-based discriminator unsupervised anomaly detection video vision transformer

来源：评论

学校读者我要写书评

暂无评论

unsupervised anomaly detection based method of risk evaluation for road traffic accident

引用

APPLIED INTELLIGENCE 2023年第1期53卷 369-384页

作者： Zhao, Chao Chang, Xiaokun Xie, Tian Fujita, Hamido Wu, Jian Shanghai Maritime Univ Sch Econ & Management Shanghai 201306 Peoples R China China Design Grp Co Ltd Nanjing Peoples R China Iwate Prefectural Univ Takizawa Iwate Japan

Elevated road plays a very important role as corridors in urban traffic network, and the occurrence of traffic accidents often causes a great impact. In that sense, we propose a unique and enhanced Autoencoder (AE) to identify elevated road traffic accident (RTA) risk based on traffic anomaly detection in an unsupervised manner. An attention mechanism is introduced to extract the traffic condition features considering traffic spatiotemporal variation characteristics. Additionally, an enhanced loss is also introduced to optimize the ability of unsupervised anomaly detection (UAD) approach to detect anomalous RTA risk and persistent anomalous traffic condition, which can significantly boost the anomaly detection performance using the contaminated traffic condition datasets. To assess the RTA risk, the evaluation mechanism and discriminant threshold are used to quantitatively analyze the detected abnormal traffic condition. Finally, experiments on real traffic datasets demonstrate the effectiveness of the model.

关键词： unsupervised anomaly detection Autoencoder Risk evaluation Road traffic accident

来源：评论

学校读者我要写书评

暂无评论

unsupervised anomaly detection Based on Minimum Spanning Tree Approximated Distance Measures and Its Application to Hydropower Turbines

引用

IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING 2019年第2期16卷 654-667页

作者： Ahmed, Imtiaz Dagnino, Aldo Ding, Yu Texas A&M Univ Dept Ind & Syst Engn College Stn TX 77843 USA ABB Informat Syst Grp Cary NC 27518 USA

Anomalies are data points or a cluster of data points that lie away from the neighboring points or clusters and are inconsistent with the overall pattern of the data. anomaly detection techniques help distinguish the anomalous observations from the regular ones, and thus provide the basis for developing a standard performance guideline for process control. The process of identifying anomalies becomes complicated in the absence of labeled training data as in supervised learning. Moreover, Euclidean distance between two points is less likely able to reflect the intrinsic structural distance imposed by the underlying manifold structure. In this paper, the authors propose a minimum spanning tree (MST)-based anomaly detection method. The merit of the method is that an MST provides a new distance measure, capable of capturing the relative connectedness of data points/clusters in a complicated manifold, and could be a better (dis)similarity metric, than the simple Euclidean distance, to identify anomalies in unsupervised learning settings. The proposed method is compared with 13 popular anomaly detection methods on 20 benchmark data sets, demonstrating a considerable improvement in its ability of identifying anomalies. Furthermore, the MST-based anomaly detection is applied to the data set from a hydropower turbine and demonstrates remarkable detection competence. Note to Practitioners-This paper is motivated by the problem of unsupervised anomaly detection in a hydropower generation plant, which operates with turbine systems that are instrumented with dozens of sensors. Each turbine has subcomponents or functional areas such as several bearing systems, a generator, and so on. Sensors collect various types of data in real time such as temperature of oil inside the bearing systems, temperature of the bearings, ambient temperature, vibrations in each functional areas, a variety of harmonics in functional areas, temperature of the coil in the generator, and many more. In tota

关键词： Geodesic distance hydropower generation plant minimum spanning tree (MST) process automation unsupervised anomaly detection

来源：评论

学校读者我要写书评

暂无评论

unsupervised anomaly detection by densely contrastive learning for time series data

引用

NEURAL NETWORKS 2023年 168卷 450-458页

作者： Zhu, Wei Li, Weijian Dorsey, E. Ray Luo, Jiebo Univ Rochester Rochester NY 14627 USA Univ Rochester Med Ctr Rochester NY 14627 USA

Time series data continuously collected by different sensors play an essential role in monitoring and predicting events in many real-world applications, and anomaly detection for time series has received increasing attention during the past decades. In this paper, we propose an anomaly detection method by densely contrasting the whole time series with its sub-sequences at different timestamps in a latent space. Our approach leverages the locality property of convolutional neural networks (CNN) and integrates position embedding to effectively capture local features for sub-sequences. Simultaneously, we employ an attention mechanism to extract global features from the entire time series. By combining these local and global features, our model is trained using both instance-level contrastive learning loss and distribution-level alignment loss. Furthermore, we introduce a reconstruction loss applied to the extracted global features to prevent the potential loss of information. To validate the efficacy of our proposed technique, we conduct experiments on publicly available time-series datasets for anomaly detection. Additionally, we evaluate our method on an in-house mobile phone dataset aimed at monitoring the status of Parkinson's disease, all within an unsupervised learning framework. Our results demonstrate the effectiveness and potential of the proposed approach in tackling anomaly detection in time series data, offering promising applications in real-world scenarios.

关键词： unsupervised anomaly detection Contrastive learning Multivariate time series Parkinson's disease

来源：评论

学校读者我要写书评

暂无评论

unsupervised anomaly detection for Liquid-Fueled Rocket Propulsion Health Monitoring

引用

JOURNAL OF AEROSPACE COMPUTING INFORMATION AND COMMUNICATION 2009年第7期6卷 464-482页

作者： Schwabacher, Mark Oza, Nikunj Matthews, Bryan NASA Ames Res Ctr Intelligent Syst Div Moffett Field CA 94035 USA Stinger Ghaffarian Technolgies Inc Intelligent Syst Div Moffett Field CA 94035 USA

This article describes the results of applying four unsupervised anomaly detection algorithms to data from two rocket propulsion testbeds. The first testbed uses historical data from the Space Shuttle Main Engine. The second testbed uses data from an experimental rocket engine test stand located at NASA Stennis Space Center. The article describes nine anomalies detected by the four algorithms. The four algorithms use four different definitions of anomalousness. Orca uses a nearest-neighbor approach, defining a point to be an anomaly if its nearest neighbors in the data space are far away from it. The Inductive Monitoring System clusters the training data, and then uses the distance to the nearest cluster as its measure of anomalousness. GritBot learns rules from the training data, and then classifies points as anomalous if they violate these rules. One-class support vector machines map the data into a high-dimensional space in which most of the normal points are on one side of a hyperplane, and then classify points on the other side of the hyperplane as anomalous. Because of these different definitions of anomalousness, different algorithms detect different anomalies. We therefore conclude that it is useful to use multiple algorithms.

关键词： unsupervised anomaly detection Liquid Fuel Rocket Space Shuttle Main Engine Gaussian Mixture Models Rocket Propulsion Stennis Space Center Space Shuttle Artificial Neural Network Temperature Sensors Vibration Sensors

来源：评论

学校读者我要写书评

暂无评论

unsupervised anomaly detection via knowledge distillation with non-directly-coupled student block fusion

引用

MACHINE VISION AND APPLICATIONS 2023年第6期34卷 1-15页

作者： Feng, Zhiyuan Chen, Ying Xie, Linbo Jiangnan Univ Key Lab Adv Proc Control Light Ind Minist Educ Wuxi 214122 Peoples R China Jiangnan Univ Engn Res Ctr Internet Things Technol Applicat Minist Educ Wuxi 214122 Peoples R China

Recently, knowledge distillation has achieved excellent results in unsupervised anomaly detection. The representation difference of anomalies between teacher and student model is an essential basis for unsupervised anomaly detection. To fully exploit the diversity of anomaly representations, a novel distillation network is proposed for unsupervised anomaly detection, consisting of a complete teacher network and a set of non-directly-coupled student blocks. Instead of taking a complete network as a student which sequentially inherits the distilled knowledge from the previous layer, the student blocks are specifically designed, which independently take features of each layer of teacher network as their input and target to recover the multi-scale representation of the teacher. For each block, an adaptive weighted multi-branch feature extraction strategy is presented to enable the blocks to better focus on key messages from the teacher model. In addition, a feature reunion technique is given during distillation to make multi-scale features more robust to noisy input. The experimental results indicate that the proposed method achieves an outstanding performance on MVTec AD dataset. Compared with the baseline method, the proposed method improves by 2.21% at ROC-AUC of image level and improves by 1.00 and 2.22% for both ROC-AUC and PRO-AUC at pixel level.

关键词： unsupervised anomaly detection Knowledge distillation Non-directly-coupled student blocks Adaptive weighted multi-branch feature extraction Feature reunion

来源：评论

学校读者我要写书评

暂无评论

unsupervised anomaly detection in IoT Systems for Smart Cities

引用

IEEE TRANSACTIONS ON NETWORK SCIENCE AND ENGINEERING 2020年第4期7卷 2231-2242页

作者： Guo, Yifan Ji, Tianxi Wang, Qianlong Yu, Lixing Min, Geyong Li, Pan Case Western Reserve Univ Dept Elect Engn Comp Sci Cleveland OH 44106 USA Univ Exeter Dept Math Coll Engn Comp SciMathPhys Sci Exeter EX4 4QF Devon England

anomaly detection is critical in the Internet of Things (IoT) systems due to its wide applications for building smart cities, such as quality control in manufacturing, intrusion detection in system security, fault detection in system monitoring. Many existing schemes are problem specific and supervised approaches, which require domain knowledge and tremendous data labeling efforts. In this paper, we investigate unsupervised anomaly detection on multidimensional time series data in IoT systems, and develops a GRU-based Gaussian Mixture VAE scheme, called GGM-VAE. In particular, we employ Gated Recurrent Unit (GRU) cells to discover the correlations among time series data, and use Gaussian Mixture priors in the latent space to characterize the multimodal data. Several previous works assume simple distributions for Gaussian Mixture priors, resulting in insufficient ability to fully capture the data patterns. To overcome this issue, we design a model selection mechanism during the training process under the guidance of Bayesian Inference Criterion (BIC) to find the model which can well estimate the distribution in the Gaussian Mixture latent space. We conduct extensive simulations on four datasets and observe that our proposed scheme outperforms the state-of-the-art anomaly detection schemes and achieves up to 47.88% improvement in F1 scores on average.

关键词： anomaly detection Time series analysis Image reconstruction Training Decoding Internet of Things Smart cities Gated Recurrent Unit (GRU) Gaussian Mixture Model (GMM) iot unsupervised anomaly detection Variational Autoencoder(VAE) smart cities

来源：评论

学校读者我要写书评

暂无评论

unsupervised anomaly detection in Stream Data with Online Evolving Spiking Neural Networks

引用

NEURAL NETWORKS 2021年 139卷 118-139页

作者： Maciag, Piotr S. Kryszkiewicz, Marzena Bembenik, Robert Lobo, Jesus L. Del Ser, Javier Warsaw Univ Technol Inst Comp Sci Nowowiejska 15-19 PL-00665 Warsaw Poland Basque Res & Technol Alliance BRTA TECNALIA Parque Tecnol BizkaiaE-700 Derio 48160 Spain Univ Basque Country UPV EHU Bilbao 48013 Spain

unsupervised anomaly discovery in stream data is a research topic with many practical applications. However, in many cases, it is not easy to collect enough training data with labeled anomalies for supervised learning of an anomaly detector in order to deploy it later for identification of real anomalies in streaming data. It is thus important to design anomalies detectors that can correctly detect anomalies without access to labeled training data. Our idea is to adapt the Online evolving Spiking Neural Network (OeSNN) classifier to the anomaly detection task. As a result, we offer an Online evolving Spiking Neural Network for unsupervised anomaly detection algorithm (OeSNN-UAD), which, unlike OeSNN, works in an unsupervised way and does not separate output neurons into disjoint decision classes. OeSNN-UAD uses our proposed new two-step anomaly detection method. Also, we derive new theoretical properties of neuronal model and input layer encoding of OeSNN, which enable more effective and efficient detection of anomalies in our OeSNN-UAD approach. The proposed OeSNNUAD detector was experimentally compared with state-of-the-art unsupervised and semi-supervised detectors of anomalies in stream data from the Numenta anomaly Benchmark and Yahoo anomaly Datasets repositories. Our approach outperforms the other solutions provided in the literature in the case of data streams from the Numenta anomaly Benchmark repository. Also, in the case of real data files of the Yahoo anomaly Benchmark repository, OeSNN-UAD outperforms other selected algorithms, whereas in the case of Yahoo anomaly Benchmark synthetic data files, it provides competitive results to the results recently reported in the literature. (C) 2021 The Author(s). Published by Elsevier Ltd.

关键词： Evolving Spiking Neural Networks Outliers detection Online learning Time series data unsupervised anomaly detection Stream data

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：