检索结果-内蒙古大学图书馆

Shape-aware contrastive deep supervision for esophageal tumor segmentation from CT scans

学校读者我要写书评

暂无评论

Shape-aware contrastive deep supervision for esophageal tumo...

IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

作者： Qiangguo Jin Hui Cui Changming Sun Jiapeng Huang Ping Xuan Yiyue Xu Linlin Wang Leilei Cao Leyi Wei Ran Su School of Software Northwestern Polytechnical University Shaanxi China Yangtze River Delta Research Institute of Northwestern Polytechnical University Taicang China Department of Computer Science and Information Technology La Trobe University Melbourne Australia CSIRO Data61 Sydney Australia Department of Computer Science School of Engineering Shantou University Guangdong China Department of Radiation Oncology Shandong Cancer Hospital and Institute Shandong First Medical University and Shandong Academy of Medical Sciences Shandong China Innovation Center of Yangtze River Delta Zhejiang University Zhejiang China School of Software Shandong University Shandong China School of Computer Software College of Intelligence and Computing Tianjin University Tianjin China

Accurate tumor segmentation is crucial for esophageal cancer radiotherapy treatment planning. The low contrast among the esophagus, tumors, and surrounding tissues, and irregular tumor shapes limit the performance of automatic segmentation methods. In this paper, we aim to exploit the irregular shapes of tumors to facilitate accurate segmentation. We propose a simple and pluggable shape-aware contrastive deep supervision network (SCDSNet) with shape-aware regularization and voxel-to-voxel contrastive deep supervision. Specifically, the shape-aware regularization with an uncertainty minimization strategy encourages the precise predictions of an additional shape-aware head. The voxel-to-voxel contrastive deep supervision enhances the multi-scale shape-tumor contrast for better voxel-to-voxel prediction of shapes. The proposed method is simple and highly pluggable, which can easily be extended to other frameworks. Further, we establish a large in-house dataset on esophageal cancer to validate the effectiveness of our proposed method. The quantitative and qualitative experimental results demonstrate the effectiveness of SCDSNet on the esophageal cancer dataset.

关键词：

CDDSA: Contrastive Domain Disentanglement and Style Augmentation for Generalizable Medical Image Segmentation

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Gu, Ran Wang, Guotai Lu, Jiangshan Zhang, Jingyang Lei, Wenhui Chen, Yinan Liao, Wenjun Zhang, Shichuan Li, Kang Metaxas, Dimitris N. Zhang, Shaoting School of Mechanical and Electrical Engineering University of Electronic Science and Technology of China Chengdu China School of Biomedical Engineering Shanghai Jiao Tong University Shanghai China School of Biomedical Engineering ShanghaiTech University Shanghai China School of Electronic Information and Electrical Engineering Shanghai Jiao Tong University Shanghai China Sensetime Research Shanghai China Department of Radiation Oncology Sichuan Cancer Hospital and Institute University of Electronic Science and Technology of China Chengdu China West China Hospital-SenseTime Joint Lab West China Biomedical Big Data Center Sichuan University Chengdu China Department of Computer Science Rutgers University PiscatawayNJ08854 United States Shanghai AI Lab Shanghai China

Generalization to previously unseen images with potential domain shifts and different styles is essential for clinically applicable medical image segmentation, and the ability to disentangle domain-specific and domain-invariant features is key for achieving Domain Generalization (DG). However, existing DG methods can hardly achieve effective disentanglement to get high generalizability. To deal with this problem, we propose an efficient Contrastive Domain Disentanglement and Style Augmentation (CDDSA) framework for generalizable medical image segmentation. First, a disentangle network is proposed to decompose an image into a domain-invariant anatomical representation and a domain-specific style code, where the former is sent to a segmentation model that is not affected by the domain shift, and the disentangle network is regularized by a decoder that combines the anatomical and style codes to reconstruct the input image. Second, to achieve better disentanglement, a contrastive loss is proposed to encourage the style codes from the same domain and different domains to be compact and divergent, respectively. Thirdly, to further improve generalizability, we propose a style augmentation method based on the disentanglement representation to synthesize images in various unseen styles with shared anatomical structures. Our method was validated on a public multi-site fundus image dataset for optic cup and disc segmentation and an in-house multi-site Nasopharyngeal Carcinoma Magnetic Resonance Image (NPC-MRI) dataset for nasopharynx Gross Tumor Volume (GTVnx) segmentation. Experimental results showed that the proposed CDDSA achieved remarkable generalizability across different domains, and it outperformed several state-of-the-art methods in domain-generalizable segmentation. Code is available at https://***/HiLab-git/DAG4MIA. © 2022, CC BY.

关键词： Magnetic resonance imaging

Leveraging Herpangina data to Enhance Hospital-level Prediction of Hand-Foot-and-Mouth Disease Admissions Using UPTST

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Yu, Guoqi Yao, Hailun Zheng, Huan Xu, Ximing The Big Data Center for Children’s Medical Care Children’s Hospital Chongqing Medical University Chongqing401122 China The School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu610000 China The National Clinical Research Center for Child Health and Disorders Ministry of Education Key Laboratory of Child Development and Disorders Chongqing401122 China The Department of Orthopedics Ministry of Education Key Laboratory of Child Development and Disorders National Clinical Research Center for Child Health and Disorders China International Science and Technology Cooperation Base of Child Development and Critical Disorders Chongqing Key Laboratory of Pediatrics Children’s Hospital Chongqing Medical University Chongqing400014 China The School of Finance Nankai University Tianjin300350 China The Big Data Center for Children’s Medical Care Children’s Hospital Chongqing Medical University Chongqing401122 China

Outbreaks of hand-foot-and-mouth disease (HFMD) have been associated with significant morbidity and, in severe cases, mortality. Accurate forecasting of daily admissions of pediatric HFMD patients i1s therefore crucial for aiding the hospital in preparing for potential outbreaks and mitigating nosocomial transmissions. To address this pressing need, we propose a novel transformer-based model with a U-net shape, utilizing the patching strategy and the joint prediction strategy that capitalizes on insights from herpangina, a disease closely correlated with HFMD. This model also integrates representation learning by introducing reconstruction loss as an auxiliary loss. The results show that our U-net Patching Time Series Transformer (UPTST) model outperforms existing approaches in both long- and short-term prediction accuracy of HFMD at hospital-level. Furthermore, the exploratory extension experiments show that the model's capabilities extend beyond prediction of infectious disease, suggesting broader applicability in various domains. Copyright © 2023, The Authors. All rights reserved.

关键词： Forecasting

Prototype Learning Guided Hybrid Network for Breast Tumor Segmentation in DCE-MRI

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Zhou, Lei Zhang, Yuzhong Zhang, Jiadong Qian, Xuejun Gong, Chen Sun, Kun Ding, Zhongxiang Wang, Xing Li, Zhenhui Liu, Zaiyi Shen, Dinggang The School of Health Science and Engineering University of Shanghai for Science and Technology Shanghai China The School of Biomedical Engineering ShanghaiTech University China The School of Computer Science and Engineering Nanjing University of Science and Technology China Ruijin Hospital Shanghai Jiaotong University School of Medicine China Department of Radiology Key Laboratory of Clinical Cancer Pharmacology and Toxicology Research of Zhejiang Province China The Shanghai General Hospital Shanghai JiaoTong University China Department of Radiology The Third Affiliated Hospital of Kunming Medical University China Department of Radiology Guangdong Provincial Peo-ple's Hospital Guangdong Academy of Medical Sciences Guangzhou510080 China Guangdong Provincial Key Laboratory of Artificial Intelligence in Medical Image Analysis and Application Guangdong Provincial People's Hospital Guangdong Academy of Medical Sciences Guangzhou510080 China School of Biomedical Engineering State Key Laboratory of Advanced Medical Materials and Devices ShanghaiTech University Shanghai201210 China Shanghai United Imaging Intelligence Co. Ltd. Shanghai200232 China Shanghai Clinical Research and Trial Center Shanghai201210 China

Automated breast tumor segmentation on the basis of dynamic contrast-enhancement magnetic resonance imaging (DCE-MRI) has shown great promise in clinical practice, particularly for identifying the presence of breast disease. However, accurate segmentation of breast tumor is a challenging task, often necessitating the development of complex networks. To strike an optimal tradeoff between computational costs and segmentation performance, we propose a hybrid network via the combination of convolution neural network (CNN) and transformer layers. Specifically, the hybrid network consists of a encoder-decoder architecture by stacking convolution and deconvolution layers. Effective 3D transformer layers are then implemented after the encoder subnetworks, to capture global dependencies between the bottleneck features. To improve the efficiency of hybrid network, two parallel encoder subnetworks are designed for the decoder and the transformer layers, respectively. To further enhance the discriminative capability of hybrid network, a prototype learning guided prediction module is proposed, where the category-specified prototypical features are calculated through online clustering. All learned prototypical features are finally combined with the features from decoder for tumor mask prediction. The experimental results on private and public DCE-MRI datasets demonstrate that the proposed hybrid network achieves superior performance than the state-of-the-art (SOTA) methods, while maintaining balance between segmentation accuracy and computation cost. Moreover, we demonstrate that automatically generated tumor masks can be effectively applied to identify HER2-positive subtype from HER2-negative subtype with the similar accuracy to the analysis based on manual tumor segmentation. The source code is available at https://***/ZhouLlab/PLHN. Copyright © 2024, The Authors. All rights reserved.

关键词： Multilayer neural networks

Energy-Constrained Computation Offloading in Space-air-Ground Integrated Networks using Distributionally Robust Optimization

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Chen, Yali ai, Bo Niu, Yong Zhang, Hongliang Han, Zhu The State Key Laboratory of Rail Traffic Control and Safety Beijing Jiaotong University Beijing100044 China The Frontiers Science Center for Smart High-Speed Railway System Beijing Jiaotong University Beijing100044 China Henan Joint International Research Laboratory of Intelligent Networking and Data Analysis Zhengzhou University Zhengzhou450001 China Research Center of Networks and Communications Peng Cheng Laboratory Shenzhen China The National Mobile Communications Research Laboratory Southeast University Nanjing211189 China The Department of Electrical Engineering Princeton University PrincetonNJ08544 United States The Department of Electrical and Computer Engineering The University of Houston HoustonTX77004 United States The Department of Computer Science and Engineering Kyung Hee University Seoul446-701 Korea Republic of

With the rapid development of connecting massive devices to the Internet, especially for remote areas without cellular network infrastructures, space-air-ground integrated networks (SAGINs) emerge and offload computation-intensive tasks. In this paper, we consider a SAGIN, where multiple low-earthorbit (LEO) satellites providing connections to the cloud server, an unmanned aerial vehicle (UAV), and nearby base stations (BSs) providing edge computing services are included. The UAV flies along a fixed trajectory to collect tasks generated by Internet of Things (IoT) devices, and forwards these tasks to a BS or the cloud server for further processing. To facilitate efficient processing, the UAV needs to decide where to offload as well as the proportion of offloaded tasks. However, in practice, due to the variability of environment and actual demand, the amount of arrival tasks is uncertain. If the deterministic optimization is utilized to develop offloading strategy, unnecessary system overhead or higher task drop rate may occur, which severely damages the system robustness. To address this issue, we characterize the uncertainty with a data-driven approach, and formulate a distributionally robust optimization problem to minimize the expected energy-constrained system latency under the worst-case probability distribution. Furthermore, the distributionally robust latency optimization algorithm is proposed to reach the suboptimal solution. Finally, we perform simulations on the realworld data set, and compare with other benchmark schemes to verify the efficiency and robustness of our proposed algorithm. Copyright © 2022, The Authors. All rights reserved.

关键词： Probability distributions

Towards adversarial patch analysis and certified defense against crowd counting

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Wu, Qiming Zou, Zhikang Zhou, Pan Ye, Xiaoqing Wang, Binghui Li, Ang School of Electronic Information and Communications Huazhong University of Science and Technology Wuhan China Baidu Inc. Shenzhen China The Hubei Engineering Research Center on Big Data Security School of Cyber Science and Engineering Huazhong University of Science and Technology Wuhan China Department of Computer Science Illinois Institute of Technology Chicago United States Department of Electrical and Computer Engineering Duke University Durham United States

Crowd counting has drawn much attention due to its importance in safety-critical surveillance systems. Especially, deep neural network (DNN) methods have significantly reduced estimation errors for crowd counting missions. Recent studies have demonstrated that DNNs are vulnerable to adversarial attacks, i.e., normal images with human-imperceptible perturbations could mislead DNNs to make false predictions. In this work, we propose a robust attack strategy called Adversarial Patch Attack with Momentum (APAM) to systematically evaluate the robustness of crowd counting models, where the attacker's goal is to create an adversarial perturbation that severely degrades their performances, thus leading to public safety accidents (e.g., stampede accidents). Especially, the proposed attack leverages the extreme-density background information of input images to generate robust adversarial patches via a series of transformations (e.g., interpolation, rotation, etc.). We observe that by perturbing less than 6% of image pixels, our attacks severely degrade the performance of crowd counting systems, both digitally and physically. To better enhance the adversarial robustness of crowd counting models, we propose the first regression model-based Randomized Ablation (RA), which is more sufficient than Adversarial Training (ADT) (Mean Absolute Error of RA is 5 lower than ADT on clean samples and 30 lower than ADT on adversarial examples). Extensive experiments on five crowd counting models demonstrate the effectiveness and generality of the proposed method. Copyright © 2021, The Authors. All rights reserved.

关键词： Network security

Learning High-Resolution and Efficient Non-local Features for Brain Glioma Segmentation in MR Images 23rd

学校读者我要写书评

暂无评论

Learning High-Resolution and Efficient Non-local Features fo...

23rd International Conference on Medical Image Computing and computer-Assisted Intervention, MICCai 2020

作者： Jia, Haozhe Xia, Yong Cai, Weidong Huang, Heng Research and Development Institute of Northwestern Polytechnical University in Shenzhen Shenzhen518057 China National Engineering Laboratory for Integrated Aero-Space-Ground-Ocean Big Data Application Technology School of Computer Science and Engineering Northwestern Polytechnical University Xian710072 China School of Computer Science University of Sydney SydneyNSW2006 Australia Department of Electrical and Computer Engineering University of Pittsburgh PittsburghPA15261 United States JD Finance America Corporation Mountain ViewCA94043 United States

ISBN: (纸本)9783030597184

Brain glioma segmentation using multi-parametric magnetic resonance (MR) imaging has significant clinical value. Although 3D convolutional neural networks (CNNs) have become increasingly prevalent in delivering this segmentation task, these models still suffer from an insufficient ability to high-resolution feature representation for small and irregular regions, limited local receptive fields, and poor long-range dependencies. In this paper, we propose a 3D High-resolution and Non-local Feature Network (HNF-Net) for brain glioma segmentation using multi-parametric MR imaging. We construct HNF-Net based mainly on the parallel multi-scale fusion (PMF) module, which helps produce strong high-resolution feature representation and aggregate multi-scale contextual information. We also introduce the expectation-maximization attention (EMA) module to HNF-Net, aiming to capture the long-range dependent contextual information and reduce the feature redundancy in a lightweight fashion. We evaluated our HNF-Net on the BraTS 2019 Challenge dataset against eight top-ranking methods listed on the challenge leaderboard. Our results suggest that the proposed HNF-Net achieves improved overall performance over these methods, and our ablation study demonstrates the effectiveness of the PMF module and EMA module. © 2020, Springer Nature Switzerland AG.

关键词： Magnetic resonance imaging

Accuracy vs. Efficiency: Achieving both Through Hardware-Aware Quantization and Reconfigurable Architecture with Mixed Precision

学校读者我要写书评

暂无评论

Accuracy vs. Efficiency: Achieving both Through Hardware-Awa...

IEEE International Conference on Big data and Cloud Computing (BdCloud)

作者： Libo Chang Shengbing Zhang Huimin Du Shiyu Wang Meikang Qiu Jihe Wang School of Computer Science Northwestern Polytechnical University and National Engineering Laboratory for Integrated Aero-Space-Ground-Ocean Big Data Application Technology Xi'an China School of Electronic Engineering Xi'an University of Posts and Telecommunication Xi'an China School of Computer Science Northwestern Polytechnical University and Engineering Research Center of Embedded System Integration Ministry of Education Xi'an China Texas A&M University-Commerce Texas USA

We propose a hardware/software co-design framework, which leverages hardware-aware quantization and a reconfigurable processor to improve the computational efficiency of convolutional neural networks (CNNs) on tiny IoT devices based on reconfigurable platforms. Firstly, we proposed a multi-objective optimization value function that can weigh accuracy, the size of CNN models, and computational delay, to improve the efficiency of the mixed- precision quantization algorithm based on deep reinforcement learning. Secondly, we propose a reconfigurable CNN processor that can adapt to the computing characteristics of various quantized CNN models, as well as a reconfigurable computing array and an on-chip elastic buffer, to improve the performance and computing efficiency on edge equipment. Finally, we demonstrate the effectiveness of the proposed co-design method through an extensive evaluation of the Ultra96-V2 platform. With respect to the well-known CNNs—VGG-16, ResNet-50, and MobileNet-V2, the experimental result shows that the throughput of 216.6 GOPS, 214.0 GOPS, and 53.6 GOPS, the computing efficiency of 0.63GOPS/DSP, 0.64GOPS/DSP, and 0.24 GOPS/DSP, respectively. In addition, achieving a better optimized trade-off between the computing efficiency and accuracy compared with the recently proposed CNN processor with fixed bit-width and mixed-precision.

关键词： Adaptation models Quantization (signal) Computational modeling Reinforcement learning Throughput System-on-chip Reconfigurable architectures

Quantum open system identification via global optimization: optimally accurate Markovian models of open systems from time-series data

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Popovych, Zakhar Jacobs, Kurt Korpas, Georgios Marecek, Jakub Bondar, Denys I. Department of Physics Tulane University New OrleansLA70118 United States DEVCOM Army Research Laboratory AdelphiMD20783 United States Department of Physics University of Massachusetts at Boston BostonMA02125 United States HSBC Lab Innovation & Ventures 8 Canada Square LondonE14 5HQ United Kingdom Department of Computer Science Czech Technical University in Prague Karlovo nam. 13 Prague 2 Czech Republic Archimedes Research Unit on AI Data Science and Algorithms Athena Research and Innovation Center Maroussi Greece

Accurate models of the dynamics of quantum circuits are essential for optimizing and advancing quantum devices. Since first-principles models of environmental noise and dissipation in real quantum systems are often unavailable, deriving accurate models from measured time-series data is critical. However, identifying open quantum systems poses significant challenges: powerful methods from systems engineering can perform poorly beyond weak damping (as we show) because they fail to incorporate essential constraints required for quantum evolution (e.g., positivity). Common methods that can include these constraints are typically multi-step, fitting linear models to physically grounded master equations, often resulting in non-convex functions in which local optimization algorithms get stuck in local extrema (as we show). In this work, we solve these problems by formulating quantum system identification directly from data as a polynomial optimization problem, enabling the use of recently developed global optimization methods. These methods are essentially guaranteed to reach global optima, allowing us for the first time to efficiently obtain the most accurate Markovian model for a given system. In addition to its practical importance, this allows us to take the error of these Markovian models as an alternative (operational) measure of the non-Markovianity of a system. We test our method with the spin-boson model — a two-level system coupled to a bath of harmonic oscillators — for which we obtain the exact evolution using matrix-product-state techniques. We show that polynomial optimization using moment/sum-of-squares approaches significantly outperforms traditional optimization algorithms, and we show that even for strong damping Lindblad-form master equations can provide accurate models of the spin-boson system (fractional errors at the level of 10−3). Using Breuer’s measure of non-Markovianity as an example we show that there can be large differences between such measur

关键词： Damping