检索结果-内蒙古大学图书馆

Mammogram mass segmentation and classification based on cross-view VAE and spatial hidden factor disentanglement

PHYSICAL AND ENGINEERING SCIENCES IN MEDICINE 2024年第1期47卷 223-238页

作者： Ma, Yingran Peng, Yanjun Shandong Univ Sci & Technol Coll Comp Sci & Engn Qingdao 266590 Peoples R China Shandong Univ Sci & Technol Shandong Prov Key Lab Wisdom Min Informat Technol Qingdao 266590 Peoples R China

Breast masses are the most important clinical findings of breast carcinomas. The mass segmentation and classification in mammograms remain a crucial yet challenging topic in computer-aided diagnosis systems, as the masses show their irregularities in shape, size and texture. In this paper, we propose a new framework for mammogram mass classification and segmentation. Specifically, to utilize the complementary information within the mammographic cross-views, cranio caudal and mediolateral oblique, a cross-view based variational autoencoder (CV-VAE) combined with a spatial hidden factor disentanglement module is presented, where the two views can be reconstructed from each other through two explicitly disentangled hidden factors: class related (specified) and background common (unspecified). Then, the specified factor is not only divided into two categories: benign and malignant by a new introduced feature pyramid networks based mass classifier, but also used to predict the mass mask label based on a U-Net-like decoder. By integrating the two complementary modules, more discriminative morphological and semantic features can be learned to solve the mass classification and segmentation problems simultaneously. The proposed method is evaluated on two most used public mammography datasets, CBIS-DDSM and INbreast, achieving the Dice similarity coefficient (DSC) of 92.46% and 93.70% for segmentation and the area under receiver operating characteristic curve (AUC) of 93.20% and 95.01% for classification, respectively. Compared with other state-of-the-art approaches, it gives competitive results.

关键词： Mammography diagnosis Cross-view mammograms variational autoencoder Hidden factor disentanglement

来源：评论

学校读者我要写书评

暂无评论

GAN-enhanced E-nose analysis: VTAAE for temporal dynamics in beef quality assessment

引用

EVOLVING SYSTEMS 2024年第6期15卷 2297-2311页

作者： Vincent, Viancy Natesan, Gobalakrishnan Loyola ICAM Coll Engn & Technol Dept Informat Technol Chennai India Sri Venkateswara Coll Engn Dept Informat Technol Chennai India

Electronic nose (E-Nose) stands out as a promising solution for the rapid detection of meat quality owing to its non-destructive and low-cost nature. As the E-nose is an essential tool for aiding quality evaluation, it is crucial to analyze the complex time sequence information produced to ensure accurate quality recognition. This research presents a novel variational Temporal Attention autoencoder with Generative Adversarial Network (VTAAE-GAN) approach to analyze time series data collected from the e-nose to evaluate beef quality. The VTAAE-GAN approach encompasses two concepts including GAN for producing synthetic time series data similar to real E-nose data and VTAAE method for capturing temporal dependencies from the complex time series data to categorize the beef quality into respective classes including good, acceptable, spoiled, and excellent. The E-Nose dataset containing time series data from eleven metal oxide semiconductor gas sensors (MOX) of twelve beef cuts is taken for beef quality evaluation. Moreover, the comprehensive experimental assessment is performed by employing existing quality detection approaches in measures of accuracy F1-score, recall, and precision. The findings suggested that the VTAAE-GAN approach by achieving an accuracy of 98.71%, recall of 98.20%, F1-score of 98.04%, and precision of 97.90% outperformed other baseline models and also evidenced its stability in freshness evaluation. Further, Maximum Mean Discrepancy, Kulback-Leibler divergence, and Wasserstein Distance metrics are used to determine the authenticity of the generated time series data. It also illustrated impressive performance with lesser variations between the real and synthetic time series data and the superior experimental outcomes reveal that our VTAAE-GAN approach is capable of learning efficient temporal features from the time series data and producing significant outcomes. Overall, the GAN incorporated for producing synthetic time series boosts the quality eval

关键词： Quality monitoring system Beef quality E-nose Time series data Machine learning GAN Attention mechanism variational autoencoder

来源：评论

学校读者我要写书评

暂无评论

An Explainable Deep Learning-Based Method for Schizophrenia Diagnosis Using Generative Data-Augmentation

引用

IEEE ACCESS 2024年 12卷 98379-98392页

作者： Saadatinia, Mehrshad Salimi-Badr, Armin Shahid Beheshti Univ Fac Comp Sci & Engn Tehran *** Iran

Schizophrenia is an example of a rare mental disorder that is challenging to diagnose using conventional methods. Deep learning methods have been extensively employed to aid in the diagnosis of schizophrenia. However, their efficacy relies heavily on data quantity, and their black-box nature raises trust concerns, especially in medical diagnosis contexts. In this study, we leverage a deep learning-based method for the automatic diagnosis of schizophrenia using EEG brain recordings. This approach utilizes generative data augmentation, a powerful technique that enhances the accuracy of the diagnosis. Additionally, our study provides a framework to use when dealing with the challenge of limited training data for the diagnosis of other potential rare mental disorders. To enable the utilization of time-frequency features, spectrograms were extracted from the raw signals. After exploring several neural network architectural setups, a proper convolutional neural network (CNN) was used for the initial diagnosis. Subsequently, using Wasserstein GAN with Gradient Penalty (WGAN-GP) and variational autoencoder (VAE), two different synthetic datasets were generated in order to augment the initial dataset and address the over-fitting issue. The augmented dataset using VAE achieved a 3.0% improvement in accuracy, reaching 99.0%, and also demonstrated faster convergence. Finally, we addressed the lack of trust in black-box models using the Local Interpretable Model-agnostic Explanations (LIME) algorithm to determine the most important superpixels (frequencies) in the diagnosis process.

关键词： Mental disorders Brain modeling Electroencephalography Accuracy Convolutional neural networks Feature extraction Data models Medical diagnosis Data augmentation Encoders schizophrenia generative data augmentation variational autoencoder generative adversarial networks explainable artificial intelligence (XAI)

来源：评论

学校读者我要写书评

暂无评论

Offline Data-Driven Adaptive Critic Design With variational Inference for Wastewater Treatment Process Control

引用

IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING 2024年第4期21卷 4987-4998页

作者： Qiao, Junfei Yang, Ruyue Wang, Ding Beijing Univ Technol Fac Informat Technol Beijing Key Lab Computat Intelligence & Intelligen Beijing Lab Smart Environm Protect Beijing 100124 Peoples R China Beijing Univ Technol Beijing Inst Artificial Intelligence Beijing 100124 Peoples R China

Wastewater treatment is indispensable to the functioning of urban society, and its optimal control has enormous social benefits. However, precise modelling of the unstable and complex treatment process is challenging yet crucial to the adaptive dynamic programming method. In this article, an adaptive critic algorithm with variational inference is designed to address the optimal control problem of nonlinear discrete-time systems, along with the convergence analysis. Based on the recorded system trajectory, the variational autoencoder is utilized to approximate the behavior policy of the offline dataset without system modelling and online interaction. Through policy iteration learning, the actor-critic structure can amend the policy generated by the variational autoencoder to achieve the optimal control objective. Simulations on a nonlinear system and the wastewater treatment process have verified that the proposed approach outperformed the behavior policy. Driven by the wastewater treatment process data derived from the incremental proportional-integral-derivative controller, the proposed approach can produce an optimal control policy of less tracking error and cost. Note to Practitioners-When dealing with an unknown system with complex dynamics, it is more feasible to improve the acceptable performance of the existing control policy based on the system's trajectory than to obtain an excelling policy. Motivated by batch reinforcement learning, learning from offline data can avoid the online interaction between the system and the adaptive dynamic programming algorithm, which could lead to exploratory errors during online learning. Specifically, using a model-free adaptive dynamic programming algorithm, the parameters of the controller are instantly updated based on the experience replay buffer sampled from the online trajectory data. However, online exploration determines the update, and there is no guarantee that the system will converge every time. As a specific typ

关键词： Adaptive dynamic programming offline reinforcement learning data-driven control variational autoencoder wastewater treatment

来源：评论

学校读者我要写书评

暂无评论

Joint Task and Data-Oriented Semantic Communications: A Deep Separate Source-Channel Coding Scheme

引用

IEEE INTERNET OF THINGS JOURNAL 2024年第2期11卷 2255-2272页

作者： Huang, Jianhao Li, Dongxu Huang, Chuan Qin, Xiaoqi Zhang, Wei Chinese Univ Hong Kong Sch Sci & Engn Shenzhen 518172 Peoples R China Chinese Univ Hong Kong Future Network Intelligence Inst Shenzhen 518172 Peoples R China Beijing Univ Posts & Telecommun State Key Lab Networking & Switching Technol Beijing 100876 Peoples R China Univ New South Wales Sch Elect Engn & Telecommun Sydney NSW 4385 Australia

Semantic communications are expected to accomplish various semantic tasks with relatively less spectrum resource by exploiting the semantic feature of source data. To simultaneously serve both the data transmission and semantic tasks, joint data compression and semantic analysis has become a pivotal issue in semantic communications. This article proposes a deep separate source-channel coding (DSSCC) framework for the joint task and data-oriented semantic communications (JTD-SCs) and utilizes the variational autoencoder approach to solve the rate-distortion problem with semantic distortion. First, by analyzing the Bayesian model of the DSSCC framework, we derive a novel rate-distortion optimization problem via the Bayesian inference approach for general data distributions and semantic tasks. Next, for a typical application of joint image transmission and classification, we combine the variational autoencoder approach with a forward adaption scheme to effectively extract image features and adaptively learn the density information of the obtained features. Finally, an iterative training algorithm is proposed to tackle the overfitting issue of deep learning models. Simulation results reveal that the proposed scheme achieves better coding gain as well as data recovery and classification performance in most scenarios, compared to the classical compression schemes and the emerging deep joint source-channel schemes.

关键词： Deep learning rate-distortion theory semantic communications separate source-channel coding variational autoencoder

来源：评论

学校读者我要写书评

暂无评论

Fund transfer fraud detection: Analyzing irregular transactions and customer with self-attention and neural networks

引用

EXPERT SYSTEMS WITH APPLICATIONS 2025年 259卷

作者： Shih, Yi-Cheng Dai, Tian-Shyr Chen, Ying-Ping Ti, Yen-Wu Wang, Wun-Hao Kuo, Yun Yang Ming Chiao Tung Univ Dept Comp Sci 1001 Daxue Rd Hsinchu 300093 Taiwan Yang Ming Chiao Tung Univ Dept Informat Management & Finance 1001 Daxue Rd Hsinchu 300093 Taiwan Natl Chengchi Univ Risk & Insurance Res Ctr 64Sec 2Zhi Nan Rd Taipei 11605 Taiwan Xiamen Ocean Vocat Coll Sch Informat Engn 61 Tiyu Rd Xiamen 361012 Fujian Peoples R China

This paper presents a method for identifying fraudulent fund transfers using real bank data, analyzing customer information, transactional activities, and customer relationships. The preprocessing step transforms high- dimensional, irregular transaction time series into regular time series, then further compresses them in a latent space using a self-attention-based autoencoder. To address the scarcity of fraudulent data samples and mitigate training issues caused by data imbalance, various deep generative models, including the conditional variational autoencoder and Wasserstein generative adversarial network, are applied to generate additional fraudulent raw data and augment fraud samples in latent space. The reparameterization trick is integrated into the encoder-decoder structure to boost the model's generative capabilities. Additionally, a Graph Neural Network (GNN) is used to model customer relationships. The proposed approach utilizes end-to-end learning, integrating the autoencoder's reconstruction loss, KL divergence loss (when reparameterization trick is applied), and classification loss for fraud detection. To optimize computational resources, neighborhood sampling for GNN is combined with mini-batch training for the autoencoder, improving both training efficiency and model reliability. Comprehensive experiments demonstrate the effectiveness of the proposed fusion network, highlighting the importance of each component and preprocessing step. For example, the areas under the precision-recall curves for fraud detection show notable improvements in our model. For suspicious transactions identified by the bank's rules, other models range from 0.66% to 22.15%, while our model reached 27%. For non-suspicious transactions, other models range from 2.53% to 22.00%, with our model achieving 22.90%. This model has potential for wider applications in anomaly detection, particularly in datasets with irregular time series and complex customer relationships.

关键词： Fraud detection Irregular time series Self-attention mechanism Graph neural networks variational autoencoder

来源：评论

学校读者我要写书评

暂无评论

VSC-ACGAN: bearing fault diagnosis model applied to imbalanced samples

引用

MEASUREMENT SCIENCE AND TECHNOLOGY 2025年第3期36卷 036212-036212页

作者： Yang, Zhenyu Mao, Runze Ye, Linchang Liu, Yun Hu, Xiaoxi Li, Yuanjiang Jiangsu Univ Sci & Technol Sch Oceanog Mengxi Rd Zhenjiang 212013 Peoples R China 711 Res Inst 400 Newton Rd Shanghai 201108 Peoples R China China Australian AI Res Inst 238 Nanxu Ave Zhenjiang 212021 Peoples R China Beijing Jiaotong Univ State Key Lab Adv Rail Autonomous Operat 3 Shangyuancun Beijing 100044 Peoples R China

The rapid development of deep learning has promoted the application of rolling bearing fault diagnosis techniques. However, in practical applications, the researchers often faces the challenge of a serious imbalance in the proportion of normal and fault states. This imbalance greatly affects the accuracy of diagnosis. Therefore, this paper proposes a novel fault diagnosis framework based on an auxiliary classifier generative adversarial network (ACGAN). Firstly, the stacked contractive autoencoder is cleverly combined with the discriminator network to improve its feature extraction capability and fault diagnosis accuracy. Subsequently, the original algorithm's focus on different types of samples is optimised to improve the generalisation of the diagnostic network. Finally, the stability of the generator network training is optimised with the help of the metric properties of the Kullback-Leibler scatter, and thus the variational stacked contractive-ACGAN model is proposed. The experimental results show that the fault diagnosis accuracy reaches 99.75 % with 200 training samples of each class on the Case Western Reserve University (CWRU) bearing dataset, which is significantly better than other algorithms. Under the same conditions, on the Jiangnan University bearing dataset, the accuracy reaches 99.25 %, which shows good generalization and provides a broad prospect for future applications.

关键词： fault diagnosis imbalanced samples auxiliary classifier generative adversarial network stacked contractive autoencoder variational autoencoder

来源：评论

学校读者我要写书评

暂无评论

State-of-health estimation and remaining useful life prediction of lithium-ion batteries using DnCNN-CNN

引用

JOURNAL OF ENERGY STORAGE 2025年 106卷

作者： Chae, Sun Geu Bae, Suk Joo Oh, Ki-Yong Hanyang Univ Dept Ind Engn Seoul South Korea Hanyang Univ Sch Mech Engn Seoul South Korea

Accurate evaluation of state-of-health (SoH) and prediction of remaining useful life (RUL) are crucial to sustain the reliability of lithium-ion batteries (LIBs) via timely maintenance actions. However, ambient noises under various operating conditions hinder accurate diagnosis of dynamic status for LIBs in real-world applications. To overcome this difficulty, an allied denoising convolutional neural network (DnCNN) and convolutional neural network (CNN) model is proposed as a new framework for estimating SoH and predicting RUL of LIBs under various operating environments. In the presence of unknown ambient noises, DnCNN is applied to improve prediction accuracy of SoH to eliminate the noises using a residual learning technique. To verify denoising abilities and resulting SoH prediction performance under real-life scenarios, multi-physics feature degradation testing data collected from custom test benches are used to evaluate its performance over competitive denoising techniques. Results from the experiments under various operating environments demonstrate that the proposed allied framework results in a higher accuracy and robustness than other state-of-the-art denoising methods in estimating SoH and predicting RUL of LIBs.

关键词： Bayesian optimization Deep learning Feature fusion Health monitoring variational autoencoder

来源：评论

学校读者我要写书评

暂无评论

An adaptive performance map generation method through shape feature fusion for the gas turbine compressor

引用

ENERGY 2025年 320卷

作者： Huang, Qinni Gu, Xiwen Zhang, Hongwei Sun, Jiahao Yang, Shixi Zhejiang Univ Sch Mech Engn State Key Lab Fluid Power & Mechatron Syst Hangzhou 310058 Peoples R China Zhejiang Univ Sch Mech Engn Key Lab Adv Mfg Technol Zhejiang Prov Hangzhou 310058 Peoples R China Hangzhou Turbine Power Grp Co Ltd Hangzhou 310022 Peoples R China

Accurate generation of the compressor performance map (CPM) is necessary for building high-precision physical models of gas turbines. The applicability and generalization ability of the current CPM generation method are relatively poor under off-design operating conditions, and the extrapolability of the generated curves is debatable. This paper proposes an adaptive performance map generation method through shape feature fusion (SFFAG) for the gas turbine compressor. A framework for the generation and optimization of the CPM curve database is developed. The nonlinear features of the CPM curves are extracted and fused by the variational autoencoder. Different CPM curves conforming to the physical shape are reconstructed to build a high-quality CPM curve database. Correction factors are introduced to adaptively correct the CPM curves by genetic algorithm. The proposed method is verified on an aircraft engine and an in-service industrial gas turbine and compared with other methods. The results show that the method can generate CPM with higher accuracy, better generalization ability and extrapolability, and better capture the nonlinear behavior of different types of gas turbines. With the continuous updating and improvement of the initial CPM curve database with richer curve shape features, the CPM curve database constructed by SFF-AG will be more complete.

关键词： Compressor performance map Adaptive generation method variational autoencoder Gas turbine Shape feature fusion

来源：评论

学校读者我要写书评

暂无评论

Ground-motion generations using Multi-label Conditional Embedding-conditional Denoising Diffusion Probabilistic Model (ML-cDDPM)

引用

SOIL DYNAMICS AND EARTHQUAKE ENGINEERING 2025年 191卷

作者： Huang, Yanwen Sun, Xiaodan You, Junjie Yang, Cheng Lu, Dagang Zhang, Jie Southwest Jiaotong Univ Dept Civil Engn Chengdu 610093 Peoples R China Natl Engn Res Ctr Technol Geol Disaster Prevent La Chengdu 610031 Peoples R China Harbin Inst Technol Dept Civil Engn Harbin 150090 Peoples R China Tongji Univ Key Lab Geotech & Underground Engn Minist Educ Shanghai 200092 Peoples R China Tongji Univ Dept Geotech Engn Shanghai 200092 Peoples R China

Current studies on ground motion prediction equations require the categorization of earthquake records by regions with different attenuation patterns. Conversely, machine learning based on entire datasets provides an opportunity to explore the factors that dominate ground motion at a site and their complex interactions. In this study, Multi-label Conditional Embedding is proposed to modify a conditional Denoising Diffusion Probabilistic Model (cDDPM) and achieve ground motion prediction of earthquake scenarios. A database was created for neural network training, consisting of 7154 horizontal ground motion records from 105 earthquakes selected from the Next Generation Attenuation (NGA)-West2 database of the Pacific Earthquake Engineering Research Center (PEER). Each record was labeled using four conditional parameters: VS30, F, MW, and Rrup. To embed multi-label and location information into the neural network, One-hot Encoding and Positional Encoding techniques were integrated. Hence, a Multi-label Conditional Embedding-conditional Denoising Diffusion Probabilistic Model (ML-cDDPM) was constructed for ground motion prediction. The model was used to simulate ground motions of past earthquakes and was compared with recorded motions. The model was also compared with two other neural-network-based prediction models. The comparisons demonstrate the reliability of ML-cDDPM in simulating ground motions for earthquake scenarios and its superiority over the other two models in representing the complexity of ground motion.

关键词： Ground motion generation Denoising diffusion probabilistic model Earthquake scenarios Multi-label conditional embedding variational autoencoder

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

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

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：